re: privacy and security: after reading about the founders, etc. I decided to trust that very little data leaves my devices.
I am considering making Rewind AI part I my permanent workflow, but I am not sure. Any experience and opinions based on other people\u2019s experiences will be appreciated."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Ask HN: Experience Using Rewind AI?"}},"_tags":["story","author_mark_l_watson","story_37782769","ask_hn"],"author":"mark_l_watson","children":[37783124,37783207],"created_at":"2023-10-05T19:05:12Z","created_at_i":1696532712,"num_comments":2,"objectID":"37782769","points":3,"story_id":37782769,"story_text":"I am using the 30 day free trial and the functionality seems very good.
re: privacy and security: after reading about the founders, etc. I decided to trust that very little data leaves my devices.
I am considering making Rewind AI part I my permanent workflow, but I am not sure. Any experience and opinions based on other people\u2019s experiences will be appreciated.","title":"Ask HN: Experience Using Rewind AI?","updated_at":"2024-09-20T15:19:27Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tikkun"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I was a paying user of Rewind.ai and loved it. They stopped development of the local one, and are doing a cloud based one instead.
What's a good MacOS local alternative that people like? I use rewind so I can take notes to myself while working by myself (automatic ongoing transcription) and so I can jog my memory about what I've been looking at (scroll back at my last 10 minutes or so of screen)."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Ask HN: Local Rewind.ai Alternative?"}},"_tags":["story","author_tikkun","story_40899644","ask_hn"],"author":"tikkun","children":[40901995],"created_at":"2024-07-07T18:59:45Z","created_at_i":1720378785,"num_comments":1,"objectID":"40899644","points":2,"story_id":40899644,"story_text":"I was a paying user of Rewind.ai and loved it. They stopped development of the local one, and are doing a cloud based one instead.
What's a good MacOS local alternative that people like? I use rewind so I can take notes to myself while working by myself (automatic ongoing transcription) and so I can jog my memory about what I've been looking at (scroll back at my last 10 minutes or so of screen).","title":"Ask HN: Local Rewind.ai Alternative?","updated_at":"2024-09-20T17:28:18Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"durron"},"title":{"fullyHighlighted":true,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Rewind AI"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/"}},"_tags":["story","author_durron","story_39461751"],"author":"durron","children":[39461779],"created_at":"2024-02-22T00:57:01Z","created_at_i":1708563421,"num_comments":1,"objectID":"39461751","points":2,"story_id":39461751,"title":"Rewind AI","updated_at":"2024-09-20T16:28:05Z","url":"https://www.rewind.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"thejarren"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Limitless AI (prev. Rewind AI) announces their Pendant and new services"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://twitter.com/dsiroker/status/1779857843895599383"}},"_tags":["story","author_thejarren","story_40042844"],"author":"thejarren","created_at":"2024-04-15T16:42:04Z","created_at_i":1713199324,"num_comments":0,"objectID":"40042844","points":2,"story_id":40042844,"title":"Limitless AI (prev. Rewind AI) announces their Pendant and new services","updated_at":"2024-09-20T16:48:48Z","url":"https://twitter.com/dsiroker/status/1779857843895599383"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"vince_liu"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"A true open source alternative to Rewind AI. It supports semantic searching and short QA on things you've seen or heard. Mostly focus on search UI.
I think building a jarvis like assistants would benefit from having a memory layer which serves as context which is exactly what Unlost is all about. Excited to see new ideas/suggestions.
Existing alternatives:\n(paid) https://www.rewind.ai/\n(open source) https://github.com/jasonjmcghee/rem"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Show HN: Unlost \u2013 Semantic search and ask your memory (Open source RewindAI)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/Vincent-liuwingsang/unlost"}},"_tags":["story","author_vince_liu","story_39561239","show_hn"],"author":"vince_liu","created_at":"2024-03-01T12:46:15Z","created_at_i":1709297175,"num_comments":0,"objectID":"39561239","points":2,"story_id":39561239,"story_text":"A true open source alternative to Rewind AI. It supports semantic searching and short QA on things you've seen or heard. Mostly focus on search UI.
I think building a jarvis like assistants would benefit from having a memory layer which serves as context which is exactly what Unlost is all about. Excited to see new ideas/suggestions.
Existing alternatives:\n(paid) https://www.rewind.ai/\n(open source) https://github.com/jasonjmcghee/rem","title":"Show HN: Unlost \u2013 Semantic search and ask your memory (Open source RewindAI)","updated_at":"2024-09-20T16:31:12Z","url":"https://github.com/Vincent-liuwingsang/unlost"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"thrownawaysz"},"story_text":{"matchLevel":"none","matchedWords":[],"value":"Are there any good alternatives?
I know about OpenRecall and ActivityWatch but what are some other options?
https://github.com/openrecall/openrecall
https://activitywatch.net/
I remember there was an older Windows desktop app too (proprietary paid) which was linked here but I forgot to save the link..."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Ask HN: Windows Recall / Rewind.ai Alternatives?"}},"_tags":["story","author_thrownawaysz","story_41484099","ask_hn"],"author":"thrownawaysz","children":[41485316,41485333],"created_at":"2024-09-08T23:20:55Z","created_at_i":1725837655,"num_comments":1,"objectID":"41484099","points":1,"story_id":41484099,"story_text":"Are there any good alternatives?
I know about OpenRecall and ActivityWatch but what are some other options?
https://github.com/openrecall/openrecall
https://activitywatch.net/
I remember there was an older Windows desktop app too (proprietary paid) which was linked here but I forgot to save the link...","title":"Ask HN: Windows Recall / Rewind.ai Alternatives?","updated_at":"2024-09-20T17:48:44Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"matteocontrini"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Rewind.ai now has a free plan"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/pricing"}},"_tags":["story","author_matteocontrini","story_37362608"],"author":"matteocontrini","created_at":"2023-09-02T15:51:41Z","created_at_i":1693669901,"num_comments":0,"objectID":"37362608","points":1,"story_id":37362608,"title":"Rewind.ai now has a free plan","updated_at":"2024-09-20T15:03:05Z","url":"https://www.rewind.ai/pricing"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"iansinnott"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Rewind.ai Review"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://blog.iansinnott.com/rewind.ai-review"}},"_tags":["story","author_iansinnott","story_35362387"],"author":"iansinnott","created_at":"2023-03-29T19:50:18Z","created_at_i":1680119418,"num_comments":0,"objectID":"35362387","points":1,"story_id":35362387,"title":"Rewind.ai Review","updated_at":"2024-09-20T13:40:08Z","url":"https://blog.iansinnott.com/rewind.ai-review"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"gtbcb"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I was considering doing this with Rewind.ai since it shut down. I've since then run across ScreenPipe, which I may use, or I may modify their open source version to make it more personalized.
I'm curious what other tools people have done this for and the most feasible and useful categories of tools for personalized AI versions."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: What paid tools have you now replaced with personalized AI-coded tools?"}},"_tags":["story","author_gtbcb","story_49184620","ask_hn"],"author":"gtbcb","children":[49188131,49191040,49192399,49192951,49195087],"created_at":"2026-08-05T15:53:33Z","created_at_i":1785945213,"num_comments":3,"objectID":"49184620","points":3,"story_id":49184620,"story_text":"I was considering doing this with Rewind.ai since it shut down. I've since then run across ScreenPipe, which I may use, or I may modify their open source version to make it more personalized.
I'm curious what other tools people have done this for and the most feasible and useful categories of tools for personalized AI versions.","title":"Ask HN: What paid tools have you now replaced with personalized AI-coded tools?","updated_at":"2026-08-06T16:04:24Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"FatihTahta"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I built an AI bot to scrape app recommendations from entrepreneurs and techies like Alex Hormozi, Noah Kagan, Tim Ferriss, Neil Patel, Pieter Levels, etc.,by analyzing their podcast transcripts and blogs.
I liked the list so I tought I can share it w reddit :
Productivity Tools:
FocusMate: Partnered productivity sessions.\nBearApp: Note-taking and planning.\nZapier: Automating repetitive tasks.\nCircle: Community and course management.\nMentio: AI for product mentions on social media.\nBusiness Intelligence Tools:
Google Tools: Simple tools for business operations.\nFivetran & Tableau: For data analysis.\nSnowflake: Storing large datasets in the cloud.\nLooker & Metabase: Data visualization.\nWebsite Tools:
Webflow: No-code website creation.\nWordPress: Hosting and managing content.\nClickFunnels: Building high-converting websites and sales funnels.\nSplunk: Site monitoring and alerting.\nSentry: Application monitoring.\nContent Tools:
Frame.io: Video collaboration and feedback.\nEpidemic Sound: Royalty-free music.\nCanva: Graphic design.\nFigma: Detailed design work.\nHemingwayApp: Improving writing clarity.\nSendFox: Email automations and newsletters.\nKlaviyo: Email marketing.\nLater: Social media post scheduling.\nTwemex: Content inspiration.\nMeeting Tools:
TidyCal: Scheduling meetings.\nZoom: Video conferencing.\nRewind.AI: Note-taking for meetings.\nLaxis & Fathom: AI meeting assistants.\nFigjam: Online whiteboard.\nSales Tools:
Dovetail: Customer feedback analysis.\nMixMax & Hubspot: Scheduling and managing customer relationships.\nClearbit: Customer data management.\nBreezeDoc: Digital document signing.\nBetterProposals: Managing proposals.\nFindThatLead & Rocketreach: Email address finding and verification.\nFinance Tools:
Ramp: Company card and expense management.\nCarta: Equity management.\nQuickBooks: Financial tracking.\nTeam Tools:
Atlassian: Project and documentation management.\nCleanShot: Screenshots and screen recording.\nBitwarden: Password management.\nSlack & Slack Bots: Communication and updates.\nInsperity: Human resources management.\nGusto: Recommended for startups.\nGreenhouse & Contra: Hiring and freelancer management."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"I Trained an AI bot to find out how to run a $100M SaaS business"}},"_tags":["story","author_FatihTahta","story_41872315","ask_hn"],"author":"FatihTahta","children":[41872931,41874797],"created_at":"2024-10-17T18:30:06Z","created_at_i":1729189806,"num_comments":2,"objectID":"41872315","points":2,"story_id":41872315,"story_text":"I built an AI bot to scrape app recommendations from entrepreneurs and techies like Alex Hormozi, Noah Kagan, Tim Ferriss, Neil Patel, Pieter Levels, etc.,by analyzing their podcast transcripts and blogs.
I liked the list so I tought I can share it w reddit :
Productivity Tools:
FocusMate: Partnered productivity sessions.\nBearApp: Note-taking and planning.\nZapier: Automating repetitive tasks.\nCircle: Community and course management.\nMentio: AI for product mentions on social media.\nBusiness Intelligence Tools:
Google Tools: Simple tools for business operations.\nFivetran & Tableau: For data analysis.\nSnowflake: Storing large datasets in the cloud.\nLooker & Metabase: Data visualization.\nWebsite Tools:
Webflow: No-code website creation.\nWordPress: Hosting and managing content.\nClickFunnels: Building high-converting websites and sales funnels.\nSplunk: Site monitoring and alerting.\nSentry: Application monitoring.\nContent Tools:
Frame.io: Video collaboration and feedback.\nEpidemic Sound: Royalty-free music.\nCanva: Graphic design.\nFigma: Detailed design work.\nHemingwayApp: Improving writing clarity.\nSendFox: Email automations and newsletters.\nKlaviyo: Email marketing.\nLater: Social media post scheduling.\nTwemex: Content inspiration.\nMeeting Tools:
TidyCal: Scheduling meetings.\nZoom: Video conferencing.\nRewind.AI: Note-taking for meetings.\nLaxis & Fathom: AI meeting assistants.\nFigjam: Online whiteboard.\nSales Tools:
Dovetail: Customer feedback analysis.\nMixMax & Hubspot: Scheduling and managing customer relationships.\nClearbit: Customer data management.\nBreezeDoc: Digital document signing.\nBetterProposals: Managing proposals.\nFindThatLead & Rocketreach: Email address finding and verification.\nFinance Tools:
Ramp: Company card and expense management.\nCarta: Equity management.\nQuickBooks: Financial tracking.\nTeam Tools:
Atlassian: Project and documentation management.\nCleanShot: Screenshots and screen recording.\nBitwarden: Password management.\nSlack & Slack Bots: Communication and updates.\nInsperity: Human resources management.\nGusto: Recommended for startups.\nGreenhouse & Contra: Hiring and freelancer management.","title":"I Trained an AI bot to find out how to run a $100M SaaS business","updated_at":"2025-08-21T19:07:33Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"vince_liu"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Pretty excited about my first show hn.
Unlost is an app that lets you semantically search your memory. Similar to rewind.ai but more focused on searching experience.
Hit me"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Unlost \u2013 Semantic search your memory"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://unlost.ai/"}},"_tags":["story","author_vince_liu","story_37622016","show_hn"],"author":"vince_liu","created_at":"2023-09-23T10:31:04Z","created_at_i":1695465064,"num_comments":0,"objectID":"37622016","points":2,"story_id":37622016,"story_text":"Pretty excited about my first show hn.
Unlost is an app that lets you semantically search your memory. Similar to rewind.ai but more focused on searching experience.
Hit me","title":"Show HN: Unlost \u2013 Semantic search your memory","updated_at":"2024-09-20T15:10:19Z","url":"https://unlost.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"inetsee"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I spotted an ad for a new product in development (Shipping Q4) that seems to offer some very interesting features. However, I had some nagging doubts, even though the product is not all that expensive ($99 + $15 shipping). Limitless appears to be the new name for "Rewind.ai", and Rewind had a product called the Rewind Pendant, that has features a lot like the new Limitless pendant, and the old product was priced at only $59.
Has anyone here looked into the Limitless AI Pendant, and would you be willing to share your thoughts?"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: Limitless: A New (to-good-to-be-true?) AI Gadget?"}},"_tags":["story","author_inetsee","story_40069141","ask_hn"],"author":"inetsee","children":[40069837],"created_at":"2024-04-17T19:32:26Z","created_at_i":1713382346,"num_comments":1,"objectID":"40069141","points":1,"story_id":40069141,"story_text":"I spotted an ad for a new product in development (Shipping Q4) that seems to offer some very interesting features. However, I had some nagging doubts, even though the product is not all that expensive ($99 + $15 shipping). Limitless appears to be the new name for "Rewind.ai", and Rewind had a product called the Rewind Pendant, that has features a lot like the new Limitless pendant, and the old product was priced at only $59.
Has anyone here looked into the Limitless AI Pendant, and would you be willing to share your thoughts?","title":"Ask HN: Limitless: A New (to-good-to-be-true?) AI Gadget?","updated_at":"2024-09-20T16:51:29Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mxstbr"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Rewind: The Search Engine for Your Life"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai"}},"_tags":["story","author_mxstbr","story_33421751"],"author":"mxstbr","children":[33422444,33422588,33422803,33422925,33422942,33422982,33423033,33423085,33423167,33423273,33423386,33423410,33423531,33423550,33423571,33423587,33423642,33423685,33423769,33423807,33423922,33424179,33424234,33424257,33424475,33424493,33424597,33424608,33424610,33424783,33424974,33425006,33425956,33426607,33432521,33432555],"created_at":"2022-11-01T14:34:08Z","created_at_i":1667313248,"num_comments":90,"objectID":"33421751","points":164,"story_id":33421751,"title":"Rewind: The Search Engine for Your Life","updated_at":"2026-04-18T10:29:22Z","url":"https://www.rewind.ai"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"davidbarker"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Rewind Pendant: a wearable that captures what you say and hear"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/pendant"}},"_tags":["story","author_davidbarker","story_37744213"],"author":"davidbarker","children":[37744391,37744443,37744450,37744463,37744480,37744495,37744501,37744517,37744522,37744524,37744525,37744582,37744586,37744606,37744613,37744626,37744640,37744681,37744696,37744697,37744704,37744714,37744727,37744736,37744780,37744786,37744787,37744818,37744967,37745510,37745530,37745770,37746191,37746639,37747531,37747681,37748627,37751033,37751525,37751557,37776405],"created_at":"2023-10-02T20:48:50Z","created_at_i":1696279730,"num_comments":122,"objectID":"37744213","points":87,"story_id":37744213,"title":"Rewind Pendant: a wearable that captures what you say and hear","updated_at":"2026-06-15T04:47:44Z","url":"https://www.rewind.ai/pendant"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dsiroker"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Rewind for iPhone"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/#iphone"}},"_tags":["story","author_dsiroker","story_36877100"],"author":"dsiroker","children":[36877101,36884148],"created_at":"2023-07-26T13:13:07Z","created_at_i":1690377187,"num_comments":2,"objectID":"36877100","points":35,"story_id":36877100,"title":"Rewind for iPhone","updated_at":"2024-09-20T14:41:25Z","url":"https://www.rewind.ai/#iphone"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dsiroker"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Rewind: The search engine for your life"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/?ref=hn"}},"_tags":["story","author_dsiroker","story_33420094"],"author":"dsiroker","children":[33420095,33428571],"created_at":"2022-11-01T12:08:23Z","created_at_i":1667304503,"num_comments":2,"objectID":"33420094","points":21,"story_id":33420094,"title":"Rewind: The search engine for your life","updated_at":"2026-04-18T10:29:22Z","url":"https://www.rewind.ai/?ref=hn"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dsiroker"},"title":{"matchLevel":"none","matchedWords":[],"value":"ChatGPT for Me"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/chatgpt-for-me"}},"_tags":["story","author_dsiroker","story_35271508"],"author":"dsiroker","children":[35271781],"created_at":"2023-03-23T07:28:48Z","created_at_i":1679556528,"num_comments":1,"objectID":"35271508","points":7,"story_id":35271508,"title":"ChatGPT for Me","updated_at":"2024-09-20T13:40:50Z","url":"https://www.rewind.ai/chatgpt-for-me"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"AlphaWeaver"},"title":{"fullyHighlighted":true,"matchLevel":"partial","matchedWords":["rewind"],"value":"Rewind"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/?2023"}},"_tags":["story","author_AlphaWeaver","story_38437289"],"author":"AlphaWeaver","created_at":"2023-11-27T19:52:44Z","created_at_i":1701114764,"num_comments":0,"objectID":"38437289","points":2,"story_id":38437289,"title":"Rewind","updated_at":"2024-09-20T15:47:45Z","url":"https://www.rewind.ai/?2023"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"iNic"},"title":{"matchLevel":"none","matchedWords":[],"value":"ChatGPT for Me"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/chatgpt-for-me"}},"_tags":["story","author_iNic","story_35290616"],"author":"iNic","created_at":"2023-03-24T15:44:47Z","created_at_i":1679672687,"num_comments":0,"objectID":"35290616","points":2,"story_id":35290616,"title":"ChatGPT for Me","updated_at":"2024-09-20T13:42:47Z","url":"https://www.rewind.ai/chatgpt-for-me"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"_xnmw"},"title":{"matchLevel":"none","matchedWords":[],"value":"ChatGPT for Me"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/chatgpt-for-me"}},"_tags":["story","author__xnmw","story_35285544"],"author":"_xnmw","created_at":"2023-03-24T05:58:25Z","created_at_i":1679637505,"num_comments":0,"objectID":"35285544","points":2,"story_id":35285544,"title":"ChatGPT for Me","updated_at":"2025-10-07T06:50:20Z","url":"https://www.rewind.ai/chatgpt-for-me"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"brettbejcek"},"title":{"matchLevel":"none","matchedWords":[],"value":"ChatGPT for Me: a copilot for your mind"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/chatgpt-for-me?ref=hn"}},"_tags":["story","author_brettbejcek","story_35281250"],"author":"brettbejcek","created_at":"2023-03-23T21:04:50Z","created_at_i":1679605490,"num_comments":0,"objectID":"35281250","points":2,"story_id":35281250,"title":"ChatGPT for Me: a copilot for your mind","updated_at":"2024-09-20T13:41:54Z","url":"https://www.rewind.ai/chatgpt-for-me?ref=hn"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"cyanf"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Ask Rewind"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.rewind.ai/ask-rewind"}},"_tags":["story","author_cyanf","story_35396859"],"author":"cyanf","created_at":"2023-04-01T03:35:52Z","created_at_i":1680320152,"num_comments":0,"objectID":"35396859","points":1,"story_id":35396859,"title":"Ask Rewind","updated_at":"2024-09-20T13:43:53Z","url":"https://www.rewind.ai/ask-rewind"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"iotambat"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"just finished an internship as a design engineer/creative tech at Tesla, was thinking of where I wanted to go next, and in my mind that's somewhere in the ai consumer/consumer hardware space. I think there's a lot of novel problems to solve in that space that haven't been tackled before, but I'm not sure what companies to look into other than:
- Humane AI\n- Rabbit \n- Rewind Pendant\n- Friend
I was thinking that if i was 10 years older I would have loved to work on Pebble (smartwatch co)
Does anyone has any other consumer ai/consumer hardware startups they recommend I look into?"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: List of consumer AI/consumer hardware startups?"}},"_tags":["story","author_iotambat","story_42966469","ask_hn"],"author":"iotambat","children":[42966498,42966505,42966777,42967113],"created_at":"2025-02-06T21:02:16Z","created_at_i":1738875736,"num_comments":5,"objectID":"42966469","points":7,"story_id":42966469,"story_text":"just finished an internship as a design engineer/creative tech at Tesla, was thinking of where I wanted to go next, and in my mind that's somewhere in the ai consumer/consumer hardware space. I think there's a lot of novel problems to solve in that space that haven't been tackled before, but I'm not sure what companies to look into other than:
- Humane AI\n- Rabbit \n- Rewind Pendant\n- Friend
I was thinking that if i was 10 years older I would have loved to work on Pebble (smartwatch co)
Does anyone has any other consumer ai/consumer hardware startups they recommend I look into?","title":"Ask HN: List of consumer AI/consumer hardware startups?","updated_at":"2025-02-20T02:56:41Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"slymersonn"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Humane is betting on the AI Pin, Rewind and Tab are betting on necklaces, etc.
In my opinion, a smartwatch would be awesome as the form factor for an "AI Partner." Glasses would also be cool, but I think it's hard to make humans want to have something on their faces for extended periods of time.
I believe that in the future, everyone will have their own personal AI, similar to how everyone has their own personal computer. The question is, what form factor makes the most sense for an "AI Partner"?
What do you think? I'm curious to hear your opinions."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: Is the smartwatch the future of AI-native devices?"}},"_tags":["story","author_slymersonn","story_38465002","ask_hn"],"author":"slymersonn","children":[38465031],"created_at":"2023-11-29T20:41:14Z","created_at_i":1701290474,"num_comments":2,"objectID":"38465002","points":1,"story_id":38465002,"story_text":"Humane is betting on the AI Pin, Rewind and Tab are betting on necklaces, etc.
In my opinion, a smartwatch would be awesome as the form factor for an "AI Partner." Glasses would also be cool, but I think it's hard to make humans want to have something on their faces for extended periods of time.
I believe that in the future, everyone will have their own personal AI, similar to how everyone has their own personal computer. The question is, what form factor makes the most sense for an "AI Partner"?
What do you think? I'm curious to hear your opinions.","title":"Ask HN: Is the smartwatch the future of AI-native devices?","updated_at":"2024-09-20T15:50:32Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"slymersonn"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I've seen a bunch of AI wearables come out, such as the Humane AI Pin, the Rewind Pendant, the Tab Necklace, etc.
The value proposition that gets thrown around with these devices is the promise of having a "second brain" there with you at all times, listening to your conversations, and gathering the context. Context being the most important part.
However, the goal for this is to be common and popular among "normies," and I'm not entirely sure if normies want to have a device that is essentially a conversation analyzer/summarizer.
Putting myself in the shoes of someone outside the tech bubble... they don't generally care about privacy, they only want something that provides utility. I say this because I know some may raise privacy concerns, but those concerns may not resonate with the average person. Therefore, trying to think like an average person, and I think what will win the market will be something that provides utility, and is a status symbol.
So this begs the question... what's the true value proposition for these devices? Is the one thing that will catch on with normies a proactive AI?"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: What's the value prop for \"AI wearables?\""}},"_tags":["story","author_slymersonn","story_39823503","ask_hn"],"author":"slymersonn","children":[39823892],"created_at":"2024-03-26T01:52:03Z","created_at_i":1711417923,"num_comments":1,"objectID":"39823503","points":1,"story_id":39823503,"story_text":"I've seen a bunch of AI wearables come out, such as the Humane AI Pin, the Rewind Pendant, the Tab Necklace, etc.
The value proposition that gets thrown around with these devices is the promise of having a "second brain" there with you at all times, listening to your conversations, and gathering the context. Context being the most important part.
However, the goal for this is to be common and popular among "normies," and I'm not entirely sure if normies want to have a device that is essentially a conversation analyzer/summarizer.
Putting myself in the shoes of someone outside the tech bubble... they don't generally care about privacy, they only want something that provides utility. I say this because I know some may raise privacy concerns, but those concerns may not resonate with the average person. Therefore, trying to think like an average person, and I think what will win the market will be something that provides utility, and is a status symbol.
So this begs the question... what's the true value proposition for these devices? Is the one thing that will catch on with normies a proactive AI?","title":"Ask HN: What's the value prop for \"AI wearables?\"","updated_at":"2024-09-20T16:39:08Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"kong_king"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I have been noticing the rise of AI wearables(Tab,Humane,Rewind) recently. However, I am wondering how these wearables store speech when their companion apps are killed(force closed)?
Do they transcribe on their MCUs itself? Or do they store it(voice/text) in a flash memory?
Or if they make a request to the server, do they assume data is always available?"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: How do AI wearables transcribe speech without their companion apps?"}},"_tags":["story","author_kong_king","story_39356431","ask_hn"],"author":"kong_king","created_at":"2024-02-13T10:43:22Z","created_at_i":1707821002,"num_comments":0,"objectID":"39356431","points":1,"story_id":39356431,"story_text":"I have been noticing the rise of AI wearables(Tab,Humane,Rewind) recently. However, I am wondering how these wearables store speech when their companion apps are killed(force closed)?
Do they transcribe on their MCUs itself? Or do they store it(voice/text) in a flash memory?
Or if they make a request to the server, do they assume data is always available?","title":"Ask HN: How do AI wearables transcribe speech without their companion apps?","updated_at":"2024-09-20T16:26:42Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"JoeOfTexas"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Hi HN, I'm Joe.
I built Authrot, because Google Docs and Word slowed to a crawl on my phone once my story passed about 25 chapters. I started with what I wanted, fast navigation between chapters, and scrolling that stays smooth however long the book gets.
Then, I wanted the story itself mapped out into a graph of connections by characters, places, and objects, all referenced to a chapter passage with context. Using AI to read each chapter, my prompts pick out these elements, and tracks how they change from chapter to chapter.
Here is a summary of features I built:
- Planner for building the story foundation, arcs and chapters
- Continuity, setups and payoffs, open threads, and plot holes flagged against the earlier fact they break
- A relationship web you can rewind to any chapter
- Chat with AI about your story to help correct issues of any kind, or even draft the whole chapter.
- Real-time document writing with collaborators, plus branches for alternate endings
- Import parsed in your browser, export to EPUB, PDF and Word
- Serial publishing on a schedule, and narration with a voice for each character
It's completely free to use, no subscriptions. You get 50 free credits to use AI. I would love feedback, especially from anyone who has written for LitRPG or fantasy."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Authrot \u2013 a writing studio that maps your characters, places and plot"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://authrot.app"}},"_tags":["story","author_JoeOfTexas","story_49978646","show_hn"],"author":"JoeOfTexas","children":[49984278,49985091],"created_at":"2026-10-06T14:01:08Z","created_at_i":1791295268,"num_comments":0,"objectID":"49978646","points":2,"story_id":49978646,"story_text":"Hi HN, I'm Joe.
I built Authrot, because Google Docs and Word slowed to a crawl on my phone once my story passed about 25 chapters. I started with what I wanted, fast navigation between chapters, and scrolling that stays smooth however long the book gets.
Then, I wanted the story itself mapped out into a graph of connections by characters, places, and objects, all referenced to a chapter passage with context. Using AI to read each chapter, my prompts pick out these elements, and tracks how they change from chapter to chapter.
Here is a summary of features I built:
- Planner for building the story foundation, arcs and chapters
- Continuity, setups and payoffs, open threads, and plot holes flagged against the earlier fact they break
- A relationship web you can rewind to any chapter
- Chat with AI about your story to help correct issues of any kind, or even draft the whole chapter.
- Real-time document writing with collaborators, plus branches for alternate endings
- Import parsed in your browser, export to EPUB, PDF and Word
- Serial publishing on a schedule, and narration with a voice for each character
It's completely free to use, no subscriptions. You get 50 free credits to use AI. I would love feedback, especially from anyone who has written for LitRPG or fantasy.","title":"Show HN: Authrot \u2013 a writing studio that maps your characters, places and plot","updated_at":"2026-10-08T00:49:52Z","url":"https://authrot.app"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"atlasunshrugged"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Companies race to make AI you can wear"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.axios.com/2023/10/04/ai-wearables-meta-humane-tab-rewind"}},"_tags":["story","author_atlasunshrugged","story_37764006"],"author":"atlasunshrugged","created_at":"2023-10-04T12:11:11Z","created_at_i":1696421471,"num_comments":0,"objectID":"37764006","points":1,"story_id":37764006,"title":"Companies race to make AI you can wear","updated_at":"2024-09-20T15:17:31Z","url":"https://www.axios.com/2023/10/04/ai-wearables-meta-humane-tab-rewind"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"danenania"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Hey HN, I'm building Plandex (https://plandex.ai), an open source, terminal-based AI coding engine for complex tasks.
I built Plandex because I was tired of copying and pasting code back and forth between ChatGPT and my projects. It can complete tasks that span multiple files and require many steps. It uses the OpenAI API with your API key (support for other models, including Claude, Gemini, and open source models is on the roadmap). You can watch a 2 minute demo here: https://player.vimeo.com/video/926634577
Here's a prompt I used to build the AWS infrastructure for Plandex Cloud (Plandex can be self-hosted or cloud-hosted): https://github.com/plandex-ai/plandex/blob/main/test/test_pr...
Something I think sets Plandex apart is a focus on working around bad outputs and iterating on tasks systematically. It's relatively easy to make a great looking demo for any tool, but the day-to-day of working with it has a lot more to do with how it handles edge cases and failures. Plandex tries to tighten the feedback loop between developer and LLM:
- Every aspect of a Plandex plan is version-controlled, from the context to the conversation itself to model settings. As soon as things start to go off the rails, you can use the `plandex rewind` command to back up and add more context or iterate on the prompt. Git-style branches allow you to test and compare multiple approaches.
- As a plan proceeds, tentative updates are accumulated in a protected sandbox (also version-controlled), preventing any wayward edits to your project files.
- The `plandex changes` command opens a diff review TUI that lets you review pending changes side-by-side like the GitHub PR review UI. Just hit the 'r' key to reject any change that doesn\u2019t look right. Once you\u2019re satisfied, either press ctrl+a from the changes TUI or run `plandex apply` to apply the changes.
- If you work on files you\u2019ve loaded into context outside of Plandex, your changes are pulled in automatically so that the model always uses the latest state of your project.
Plandex makes it easy to load files and directories in the terminal. You can load multiple paths:
plandex load components/some-component.ts lib/api.ts ../sibling-dir/another-file.ts\n plandex load src/lib -r\n plandex load src/**/*.{ts,tsx}\n plandex load src --tree \n plandex load https://react.dev/reference/react/hooks\n cargo test | plandex load\n plandex tell -f "prompts/stripe/add-webhooks.txt"\n plandex tell\n plandex tell "set the header's background to #222 and text to white"\n plandex tell "write tests for all functions in lib/math/math.go. put them in lib/math_tests." --bg \n plandex ps\n plandex connect\nPlandex is written in Go and is statically compiled, so it runs from a single small binary with no dependencies on any package managers or language runtimes. There\u2019s a 1-line quick install:
curl -sL https://plandex.ai/install.sh | bash\nI built Plandex because I was tired of copying and pasting code back and forth between ChatGPT and my projects. It can complete tasks that span multiple files and require many steps. It uses the OpenAI API with your API key (support for other models, including Claude, Gemini, and open source models is on the roadmap). You can watch a 2 minute demo here: https://player.vimeo.com/video/926634577
Here's a prompt I used to build the AWS infrastructure for Plandex Cloud (Plandex can be self-hosted or cloud-hosted): https://github.com/plandex-ai/plandex/blob/main/test/test_pr...
Something I think sets Plandex apart is a focus on working around bad outputs and iterating on tasks systematically. It's relatively easy to make a great looking demo for any tool, but the day-to-day of working with it has a lot more to do with how it handles edge cases and failures. Plandex tries to tighten the feedback loop between developer and LLM:
- Every aspect of a Plandex plan is version-controlled, from the context to the conversation itself to model settings. As soon as things start to go off the rails, you can use the `plandex rewind` command to back up and add more context or iterate on the prompt. Git-style branches allow you to test and compare multiple approaches.
- As a plan proceeds, tentative updates are accumulated in a protected sandbox (also version-controlled), preventing any wayward edits to your project files.
- The `plandex changes` command opens a diff review TUI that lets you review pending changes side-by-side like the GitHub PR review UI. Just hit the 'r' key to reject any change that doesn\u2019t look right. Once you\u2019re satisfied, either press ctrl+a from the changes TUI or run `plandex apply` to apply the changes.
- If you work on files you\u2019ve loaded into context outside of Plandex, your changes are pulled in automatically so that the model always uses the latest state of your project.
Plandex makes it easy to load files and directories in the terminal. You can load multiple paths:
plandex load components/some-component.ts lib/api.ts ../sibling-dir/another-file.ts\n plandex load src/lib -r\n plandex load src/**/*.{ts,tsx}\n plandex load src --tree \n plandex load https://react.dev/reference/react/hooks\n cargo test | plandex load\n plandex tell -f "prompts/stripe/add-webhooks.txt"\n plandex tell\n plandex tell "set the header's background to #222 and text to white"\n plandex tell "write tests for all functions in lib/math/math.go. put them in lib/math_tests." --bg \n plandex ps\n plandex connect\nPlandex is written in Go and is statically compiled, so it runs from a single small binary with no dependencies on any package managers or language runtimes. There\u2019s a 1-line quick install:
curl -sL https://plandex.ai/install.sh | bash\nYou can watch a 2 minute demo of Plandex in action here: https://www.youtube.com/watch?v=SFSu2vNmlLk
And here\u2019s more of a tutorial style demo showing how Plandex can automatically debug a browser application: https://www.youtube.com/watch?v=g-_76U_nK0Y.
I launched Plandex v1 here on HN a little less than a year ago (https://news.ycombinator.com/item?id=39918500).
Now I\u2019m launching a major update, Plandex v2, which is the result of 8 months of heads down work, and is in effect a whole new project/product.
In short, Plandex is now a top-tier coding agent with fully autonomous capabilities. It combines models from Anthropic, OpenAI, and Google to achieve better results, more reliable agent behavior, better cost efficiency, and better performance than is possible by using only a single provider\u2019s models.
I believe it is now one of the best tools available for working on large tasks in real world codebases with AI. It has an effective context window of 2M tokens, and can index projects of 20M tokens and beyond using tree-sitter project maps (30+ languages are supported). It can effectively find relevant context in massive million-line projects like SQLite, Redis, and Git.
A bit more on some of Plandex\u2019s key features:
- Plandex has a built-in diff review sandbox that helps you get the benefits of AI without leaving behind a mess in your project. By default, all changes accumulate in the sandbox until you approve them. The sandbox is version-controlled. You can rewind it to any previous point, and you can also create branches to try out alternative approaches.
- It offers a \u2018full auto mode\u2019 that can complete large tasks autonomously end-to-end, including high level planning, context loading, detailed planning, implementation, command execution (for dependencies, builds, tests, etc.), and debugging.
- The autonomy level is highly configurable. You can move up and down the ladder of autonomy depending on the task, your comfort level, and how you weigh cost optimization vs. effort and results.
- Models and model settings are also very configurable. There are built-in models and model packs for different use cases. You can also add custom models and model packs, and customize model settings like temperature or top-p. All model changes are version controlled, so you can use branches to try out the same task with different models. The newly released OpenAI models and the paid Gemini 2.5 Pro model will be integrated in the default model pack soon.
- It can be easily self-hosted, including a \u2018local mode\u2019 for a very fast local single-user setup with Docker.
- Cloud hosting is also available for added convenience with a couple of subscription tiers: an \u2018Integrated Models\u2019 mode that requires no other accounts or API keys and allows you to manage billing/budgeting/spending alerts and track usage centrally, and a \u2018BYO API Key\u2019 mode that allows you to use your own OpenAI/OpenRouter accounts.
I\u2019d love to get more HNers in the Plandex Discord (https://discord.gg/plandex-ai). Please join and say hi!
And of course I\u2019d love to hear your feedback, whether positive or negative. Thanks so much!"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Plandex v2 \u2013 open source AI coding agent for large projects and tasks"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://github.com/plandex-ai/plandex"}},"_tags":["story","author_danenania","story_43710576","show_hn"],"author":"danenania","children":[43710740,43710745,43710777,43710797,43710811,43710837,43711128,43711522,43712711,43712909,43713099,43713520,43713632,43713977,43714056,43715355,43715766,43717614,43717668,43718142,43719079,43719255,43719486,43743767],"created_at":"2025-04-16T21:26:42Z","created_at_i":1744838802,"num_comments":81,"objectID":"43710576","points":257,"story_id":43710576,"story_text":"Hey HN! I\u2019m Dane, the creator of Plandex (https://github.com/plandex-ai/plandex), an open source AI coding agent focused especially on tackling large tasks in real world software projects.
You can watch a 2 minute demo of Plandex in action here: https://www.youtube.com/watch?v=SFSu2vNmlLk
And here\u2019s more of a tutorial style demo showing how Plandex can automatically debug a browser application: https://www.youtube.com/watch?v=g-_76U_nK0Y.
I launched Plandex v1 here on HN a little less than a year ago (https://news.ycombinator.com/item?id=39918500).
Now I\u2019m launching a major update, Plandex v2, which is the result of 8 months of heads down work, and is in effect a whole new project/product.
In short, Plandex is now a top-tier coding agent with fully autonomous capabilities. It combines models from Anthropic, OpenAI, and Google to achieve better results, more reliable agent behavior, better cost efficiency, and better performance than is possible by using only a single provider\u2019s models.
I believe it is now one of the best tools available for working on large tasks in real world codebases with AI. It has an effective context window of 2M tokens, and can index projects of 20M tokens and beyond using tree-sitter project maps (30+ languages are supported). It can effectively find relevant context in massive million-line projects like SQLite, Redis, and Git.
A bit more on some of Plandex\u2019s key features:
- Plandex has a built-in diff review sandbox that helps you get the benefits of AI without leaving behind a mess in your project. By default, all changes accumulate in the sandbox until you approve them. The sandbox is version-controlled. You can rewind it to any previous point, and you can also create branches to try out alternative approaches.
- It offers a \u2018full auto mode\u2019 that can complete large tasks autonomously end-to-end, including high level planning, context loading, detailed planning, implementation, command execution (for dependencies, builds, tests, etc.), and debugging.
- The autonomy level is highly configurable. You can move up and down the ladder of autonomy depending on the task, your comfort level, and how you weigh cost optimization vs. effort and results.
- Models and model settings are also very configurable. There are built-in models and model packs for different use cases. You can also add custom models and model packs, and customize model settings like temperature or top-p. All model changes are version controlled, so you can use branches to try out the same task with different models. The newly released OpenAI models and the paid Gemini 2.5 Pro model will be integrated in the default model pack soon.
- It can be easily self-hosted, including a \u2018local mode\u2019 for a very fast local single-user setup with Docker.
- Cloud hosting is also available for added convenience with a couple of subscription tiers: an \u2018Integrated Models\u2019 mode that requires no other accounts or API keys and allows you to manage billing/budgeting/spending alerts and track usage centrally, and a \u2018BYO API Key\u2019 mode that allows you to use your own OpenAI/OpenRouter accounts.
I\u2019d love to get more HNers in the Plandex Discord (https://discord.gg/plandex-ai). Please join and say hi!
And of course I\u2019d love to hear your feedback, whether positive or negative. Thanks so much!","title":"Show HN: Plandex v2 \u2013 open source AI coding agent for large projects and tasks","updated_at":"2025-06-14T00:05:39Z","url":"https://github.com/plandex-ai/plandex"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"doshay"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"hi guys.\nbeen working on something i think is fundamentally missing in today's workflow with ai agents.
vcs.
i find myself struggling with questions that agents can't answer like "why did you do it?", "when did u delete this folder? why?", etc. or trying to /rewind (after a /compact...) or basically `bisect` to find when and why something was done by the agent in the current / previous session.
just like git did for code, i think we are the same core capabilities with ai agents
so...
i developed an open source solution for that (currently supporting claude code)
would love to get feedback, contribution or maybe other ideas or solutions you find for those problems."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Git for AI Agents"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/regent-vcs/re_gent"}},"_tags":["story","author_doshay","story_48063548","show_hn"],"author":"doshay","children":[48063730,48064261,48064317,48064772,48064799,48064856,48064859,48064864,48064990,48065047,48065065,48065130,48065150,48065166,48065312,48065412,48065446,48065453,48065680,48065810,48065840,48065849,48066028,48066043,48066675,48066677,48066733,48066993,48067016,48067075,48067135,48067158,48067231,48067843,48068703,48069543,48070680,48070892,48071427,48072339,48072996,48073190,48075035,48075926,48077768,48085539,48091730,48092546,48095948,48125574,48130735,48132116,48182059],"created_at":"2026-05-08T14:15:36Z","created_at_i":1778249736,"num_comments":67,"objectID":"48063548","points":129,"story_id":48063548,"story_text":"hi guys.\nbeen working on something i think is fundamentally missing in today's workflow with ai agents.
vcs.
i find myself struggling with questions that agents can't answer like "why did you do it?", "when did u delete this folder? why?", etc. or trying to /rewind (after a /compact...) or basically `bisect` to find when and why something was done by the agent in the current / previous session.
just like git did for code, i think we are the same core capabilities with ai agents
so...
i developed an open source solution for that (currently supporting claude code)
would love to get feedback, contribution or maybe other ideas or solutions you find for those problems.","title":"Show HN: Git for AI Agents","updated_at":"2026-09-21T12:15:13Z","url":"https://github.com/regent-vcs/re_gent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ofabioroma"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Hey HN! I'm Fabio and I built UltraContext, a simple context API for AI agents with automatic versioning.
After two years building AI agents in production, I experienced firsthand how frustrating it is to manage context at scale. Storing messages, iterating system prompts, debugging behavior and multi-agent patterns\u2014all while keeping track of everything without breaking anything. It was driving me insane.
So I built UltraContext. The mental model is git for context:
- Updates and deletes automatically create versions (history is never lost)
- Replay state at any point
The API is 5 methods:
uc.create() // new context (can fork from existing)\n uc.append() // add message\n uc.get() // retrieve by version, timestamp, or index\n uc.update() // edit message \u2192 creates version\n uc.delete() // remove message \u2192 creates version\nWhat it's for:
- Persisting conversation state across sessions
- Debugging agent behavior (rewind to decision point)
- Forking contexts to test different flows
- Audit trails without building audit infrastructure
- Multi-agent and sub-agent patterns
What it's NOT:
- Not a memory/RAG system (no semantic search)
- Not a vector database
- Not an Orchestration/LLM framework
UltraContext handles versioning, branching, history. You get time-travel with one line.
Docs: https://ultracontext.ai/docs
Early access: https://ultracontext.ai
Would love feedback! Especially from anyone who's rolled their own context engineering and can tell me what I'm missing."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: UltraContext \u2013 A simple context API for AI agents with auto-versioning"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://ultracontext.ai/"}},"_tags":["story","author_ofabioroma","story_46706442","show_hn"],"author":"ofabioroma","children":[46706459,46709451,46712219,46719034],"created_at":"2026-01-21T14:48:10Z","created_at_i":1769006890,"num_comments":21,"objectID":"46706442","points":21,"story_id":46706442,"story_text":"Hey HN! I'm Fabio and I built UltraContext, a simple context API for AI agents with automatic versioning.
After two years building AI agents in production, I experienced firsthand how frustrating it is to manage context at scale. Storing messages, iterating system prompts, debugging behavior and multi-agent patterns\u2014all while keeping track of everything without breaking anything. It was driving me insane.
So I built UltraContext. The mental model is git for context:
- Updates and deletes automatically create versions (history is never lost)
- Replay state at any point
The API is 5 methods:
uc.create() // new context (can fork from existing)\n uc.append() // add message\n uc.get() // retrieve by version, timestamp, or index\n uc.update() // edit message \u2192 creates version\n uc.delete() // remove message \u2192 creates version\nWhat it's for:
- Persisting conversation state across sessions
- Debugging agent behavior (rewind to decision point)
- Forking contexts to test different flows
- Audit trails without building audit infrastructure
- Multi-agent and sub-agent patterns
What it's NOT:
- Not a memory/RAG system (no semantic search)
- Not a vector database
- Not an Orchestration/LLM framework
UltraContext handles versioning, branching, history. You get time-travel with one line.
Docs: https://ultracontext.ai/docs
Early access: https://ultracontext.ai
Would love feedback! Especially from anyone who's rolled their own context engineering and can tell me what I'm missing.","title":"Show HN: UltraContext \u2013 A simple context API for AI agents with auto-versioning","updated_at":"2026-03-05T23:22:24Z","url":"https://ultracontext.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"gloxkiqcza"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Mario Kart Wii recompiled for PC using AI, with 4K potential and uncapped FPS"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"https://www.tomshardware.com/video-games/retro-gaming/mario-kart-wii-recompiled-for-pc-using-ai-with-4k-potential-and-uncapped-frame-rates-first-static-recompilation-of-a-wii-game-supports-over-200-tracks-thanks-to-retro-rewind-compatibility"}},"_tags":["story","author_gloxkiqcza","story_48945476"],"author":"gloxkiqcza","children":[48947179],"created_at":"2026-07-17T10:16:54Z","created_at_i":1784283414,"num_comments":2,"objectID":"48945476","points":10,"story_id":48945476,"title":"Mario Kart Wii recompiled for PC using AI, with 4K potential and uncapped FPS","updated_at":"2026-07-18T14:25:45Z","url":"https://www.tomshardware.com/video-games/retro-gaming/mario-kart-wii-recompiled-for-pc-using-ai-with-4k-potential-and-uncapped-frame-rates-first-static-recompilation-of-a-wii-game-supports-over-200-tracks-thanks-to-retro-rewind-compatibility"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"lezuber"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Limitless (Rewind) Aquired by Meta"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://www.limitless.ai/"}},"_tags":["story","author_lezuber","story_46170986"],"author":"lezuber","children":[46171288,46171441],"created_at":"2025-12-06T05:31:32Z","created_at_i":1764999092,"num_comments":2,"objectID":"46170986","points":5,"story_id":46170986,"title":"Limitless (Rewind) Aquired by Meta","updated_at":"2026-03-05T23:08:43Z","url":"https://www.limitless.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"doshay"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"hi guys.
been working on something i think is fundamentally missing in today's workflow with ai agents.
vcs.
i find myself struggling with questions that agents can't answer like "why did you do it?", "when did u delete this folder? why?", etc. \nor trying to /rewind (after a /compact...) or\nbasically `bisect` to find when and why something was done by the agent in the current / previous session.
just like git did for code, i think we are the same core capabilities with ai agents
so...
i developed an open source solution for that (currently supporting claude code)
would love to get feedback, contribution or maybe other ideas or solutions you find for those problems."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Git for AI Agents"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/regent-vcs/re_gent"}},"_tags":["story","author_doshay","story_48049431","show_hn"],"author":"doshay","children":[48049432],"created_at":"2026-05-07T13:49:05Z","created_at_i":1778161745,"num_comments":0,"objectID":"48049431","points":5,"story_id":48049431,"story_text":"hi guys.
been working on something i think is fundamentally missing in today's workflow with ai agents.
vcs.
i find myself struggling with questions that agents can't answer like "why did you do it?", "when did u delete this folder? why?", etc. \nor trying to /rewind (after a /compact...) or\nbasically `bisect` to find when and why something was done by the agent in the current / previous session.
just like git did for code, i think we are the same core capabilities with ai agents
so...
i developed an open source solution for that (currently supporting claude code)
would love to get feedback, contribution or maybe other ideas or solutions you find for those problems.","title":"Show HN: Git for AI Agents","updated_at":"2026-05-10T03:13:38Z","url":"https://github.com/regent-vcs/re_gent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"slymersonn"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Over the past few months, we've seen a couple of consumer hardware startups come out with their take on what an "AI wearable" will look like. For example, the Humane Pin, Rewind Pendant, Brilliant Labs Glasses, Avi's Tab (Necklace), etc.
However, no one has tried making an AI wearable wrist device that isn't a smartwatch. I'm envisioning a wristband with no screen that is used as a device that passively listens to your conversations and gathers context (but doesn't store your recordings). I imagine this being paired with your phone so that the AI can proactively send you notifications regarding what you've talked about, almost like having a second companion always there, ready to help. This would be super cool to have as a student in a classroom because then I can have my personal AI help me study after!
This sounds cool to me, but I'm not an engineer, so I don't know why no one has attempted this. What are the hardware or software blockers preventing this from happening? Has it not been done due to engineering constraints or is it because it's a stupid idea?"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: Why hasn't anyone tried building an AI wearable wristband?"}},"_tags":["story","author_slymersonn","story_39543374","ask_hn"],"author":"slymersonn","children":[39543458,39543536,39543582,39543727,39544460,39544595],"created_at":"2024-02-28T21:02:38Z","created_at_i":1709154158,"num_comments":19,"objectID":"39543374","points":4,"story_id":39543374,"story_text":"Over the past few months, we've seen a couple of consumer hardware startups come out with their take on what an "AI wearable" will look like. For example, the Humane Pin, Rewind Pendant, Brilliant Labs Glasses, Avi's Tab (Necklace), etc.
However, no one has tried making an AI wearable wrist device that isn't a smartwatch. I'm envisioning a wristband with no screen that is used as a device that passively listens to your conversations and gathers context (but doesn't store your recordings). I imagine this being paired with your phone so that the AI can proactively send you notifications regarding what you've talked about, almost like having a second companion always there, ready to help. This would be super cool to have as a student in a classroom because then I can have my personal AI help me study after!
This sounds cool to me, but I'm not an engineer, so I don't know why no one has attempted this. What are the hardware or software blockers preventing this from happening? Has it not been done due to engineering constraints or is it because it's a stupid idea?","title":"Ask HN: Why hasn't anyone tried building an AI wearable wristband?","updated_at":"2024-09-20T16:29:09Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"BlueHotDog2"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Every agent framework gives you logs(each its own flavour of logs). Unstructured text. Maybe some spans if you're lucky. When your agent breaks something, you get to grep through a wall of output in some proprietery system.
Why can't i easily see what the agent did to produce the PR? why can't i restart a different agent from a state?
I got tired of this. agd is a content-addressed object store and causal DAG for recording agent behavior. \nIt works like git (blobs, trees, refs, immutable objects identified by hash), but the object types are specific to LLM interactions: messages with typed content parts, tool definitions, and workspace snapshots that form a causal graph.
The core idea is stolen from git: the data model is the API. You interact with objects directly. Everything is immutable, everything has a hash.
An "action" in the DAG records: what messages the agent saw (observed state), what it produced (produced state), which tools were available, which model was used, and what caused it (parent actions).
Two states per action, both full snapshots, not deltas. You diff them to see what changed.\nWhat you get:\n- agd log --session s1 shows one conversation's full causal chain\n- agd show <action> --produced --expand shows the exact prompt and tool calls\n- agd diff between two actions compares messages and workspace files\n- agd rewind moves a session back to a previous point (old actions stay in the store)\n- agd replay reconstructs the exact input state and reruns an action
It integrates as middleware/plugin. Wraps your existing LLM calls, records before/after state, doesn't require rewriting your agent code. The session ref (like a git branch pointer) auto-advances on each action, so parent tracking is a single ref read.
Written in Zig. Most of the code was written with heavy AI assistance. The store is append-only loose files, like git's object directory. Write path targets low single-digit ms per action with batched fsync. Sessions can be bundled and published to a remote for sharing and viewing(working on a POC of this, have some core questions)
This is pre-alpha. The object model and write path work. Workspace capture, session sharing, and a Phoenix LiveView web viewer are functional.
Plenty is still missing: resume-from-any-point, proper diffing, the replay command. The on-disk format will probably change. I wouldn't depend on it for anything you care about yet.
What it does not do: orchestrate agents, make agents smarter, stream in real time, or replace your framework.
Looking for feedback, thoughts, contributors
Repo: https://github.com/frontman-ai/agd"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Agd \u2013 a content-addressed DAG for tracking what AI agents do"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://github.com/frontman-ai/agd"}},"_tags":["story","author_BlueHotDog2","story_47224560","show_hn"],"author":"BlueHotDog2","children":[47231149],"created_at":"2026-03-02T21:42:40Z","created_at_i":1772487760,"num_comments":2,"objectID":"47224560","points":4,"story_id":47224560,"story_text":"Every agent framework gives you logs(each its own flavour of logs). Unstructured text. Maybe some spans if you're lucky. When your agent breaks something, you get to grep through a wall of output in some proprietery system.
Why can't i easily see what the agent did to produce the PR? why can't i restart a different agent from a state?
I got tired of this. agd is a content-addressed object store and causal DAG for recording agent behavior. \nIt works like git (blobs, trees, refs, immutable objects identified by hash), but the object types are specific to LLM interactions: messages with typed content parts, tool definitions, and workspace snapshots that form a causal graph.
The core idea is stolen from git: the data model is the API. You interact with objects directly. Everything is immutable, everything has a hash.
An "action" in the DAG records: what messages the agent saw (observed state), what it produced (produced state), which tools were available, which model was used, and what caused it (parent actions).
Two states per action, both full snapshots, not deltas. You diff them to see what changed.\nWhat you get:\n- agd log --session s1 shows one conversation's full causal chain\n- agd show <action> --produced --expand shows the exact prompt and tool calls\n- agd diff between two actions compares messages and workspace files\n- agd rewind moves a session back to a previous point (old actions stay in the store)\n- agd replay reconstructs the exact input state and reruns an action
It integrates as middleware/plugin. Wraps your existing LLM calls, records before/after state, doesn't require rewriting your agent code. The session ref (like a git branch pointer) auto-advances on each action, so parent tracking is a single ref read.
Written in Zig. Most of the code was written with heavy AI assistance. The store is append-only loose files, like git's object directory. Write path targets low single-digit ms per action with batched fsync. Sessions can be bundled and published to a remote for sharing and viewing(working on a POC of this, have some core questions)
This is pre-alpha. The object model and write path work. Workspace capture, session sharing, and a Phoenix LiveView web viewer are functional.
Plenty is still missing: resume-from-any-point, proper diffing, the replay command. The on-disk format will probably change. I wouldn't depend on it for anything you care about yet.
What it does not do: orchestrate agents, make agents smarter, stream in real time, or replace your framework.
Looking for feedback, thoughts, contributors
Repo: https://github.com/frontman-ai/agd","title":"Show HN: Agd \u2013 a content-addressed DAG for tracking what AI agents do","updated_at":"2026-03-05T23:40:07Z","url":"https://github.com/frontman-ai/agd"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"DanielHu87"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Hi HN \u2014 I\u2019m Ethan, a solo developer. I built DubTab because I spend a lot of time in meetings and watching videos in languages I\u2019m not fluent in, and subtitles alone don\u2019t always keep up (especially when the speaker is fast).
DubTab is a Chrome/Edge extension that listens to the audio of your current tab and gives you:
1.Live translated subtitles (optional bilingual mode)\n2.Optional AI dubbing with a natural-sounding voice \u2014 so you can follow by listening, not just reading
The goal is simple: make it easier to understand live audio in another language in real time, without downloading files or doing an upload-and-wait workflow.
How you\u2019d use it
1.Open a video call / livestream / lecture / any tab with audio\n2.Start DubTab\n3.Choose target language (and source language if you know it)\n4.Use subtitles only, or turn on natural AI dubbing and adjust the audio mix (keep original, or duck it)
What it\u2019s good for\n1.Following cross-language meetings/classes when you\u2019re tired of staring at subtitles\n2.Watching live content where you can\u2019t pause/rewind constantly\n3.Language learners who want bilingual captions to sanity-check meaning\n4.Keeping up with live news streams on YouTube when events are unfolding in real time (e.g., breaking international updates like U.S./Iran/Israel-related developments)
Link: https://dubtab.com
I\u2019ll be in the comments and happy to share implementation details if anyone\u2019s curious."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: DubTab \u2013 Live AI Dubbing in the Browser (Meet/YouTube/Twitch/etc.)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://dubtab.com/"}},"_tags":["story","author_DanielHu87","story_47241606","show_hn"],"author":"DanielHu87","children":[47241616],"created_at":"2026-03-04T01:04:04Z","created_at_i":1772586244,"num_comments":1,"objectID":"47241606","points":4,"story_id":47241606,"story_text":"Hi HN \u2014 I\u2019m Ethan, a solo developer. I built DubTab because I spend a lot of time in meetings and watching videos in languages I\u2019m not fluent in, and subtitles alone don\u2019t always keep up (especially when the speaker is fast).
DubTab is a Chrome/Edge extension that listens to the audio of your current tab and gives you:
1.Live translated subtitles (optional bilingual mode)\n2.Optional AI dubbing with a natural-sounding voice \u2014 so you can follow by listening, not just reading
The goal is simple: make it easier to understand live audio in another language in real time, without downloading files or doing an upload-and-wait workflow.
How you\u2019d use it
1.Open a video call / livestream / lecture / any tab with audio\n2.Start DubTab\n3.Choose target language (and source language if you know it)\n4.Use subtitles only, or turn on natural AI dubbing and adjust the audio mix (keep original, or duck it)
What it\u2019s good for\n1.Following cross-language meetings/classes when you\u2019re tired of staring at subtitles\n2.Watching live content where you can\u2019t pause/rewind constantly\n3.Language learners who want bilingual captions to sanity-check meaning\n4.Keeping up with live news streams on YouTube when events are unfolding in real time (e.g., breaking international updates like U.S./Iran/Israel-related developments)
Link: https://dubtab.com
I\u2019ll be in the comments and happy to share implementation details if anyone\u2019s curious.","title":"Show HN: DubTab \u2013 Live AI Dubbing in the Browser (Meet/YouTube/Twitch/etc.)","updated_at":"2026-03-05T23:41:03Z","url":"https://dubtab.com/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"danenania"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Hey HN! I\u2019m Dane, the creator of Plandex (https://github.com/plandex-ai/plandex), an open source AI coding agent focused especially on tackling large tasks in real world software projects.
You can watch a 2 minute demo of Plandex in action here: https://www.youtube.com/watch?v=SFSu2vNmlLk
And here\u2019s more of a tutorial style demo showing how Plandex can automatically debug a browser app: https://www.youtube.com/watch?v=VCegxOCAPq0
I launched Plandex v1 here on HN a little less than a year ago (https://news.ycombinator.com/item?id=39918500). It got a positive reception overall (plus some skepticism, which is fair enough), and a lot of early users and contributors. HNers helped to push it to the top repo on GitHub trending, where it remained for a few days, racking up over 8k stars.
Now I\u2019m launching a major update, Plandex v2, which is the result of 8 months of heads down work, and is in effect a whole new project/product.
In short, Plandex is now a top-tier coding agent with fully autonomous capabilities.
It combines models from Anthropic, OpenAI, and Google to achieve better results, more reliable agent behavior, better cost efficiency, and better performance than is possible by using only a single provider\u2019s models.
I believe it is now one of the best tools available for working on large tasks in real world codebases with AI. It has an effective context window of 2M tokens, and can index projects of 20M tokens and beyond using tree-sitter project maps (30+ languages are supported). It can effectively find relevant context in massive million-line projects like SQLite, Redis, and Git.
There are also some major UX improvements. Plandex v1 offered just a git-like CLI interface. You\u2019d send prompts via the `plandex tell` command, either passing a string, a prompt file path, or popping up vim/nano.
While that CLI interface still exists, there\u2019s now a much more intuitive chat-like REPL experience that gives you fuzzy autocomplete for commands and file paths. This is the new default/recommended way to use Plandex, though having the CLI interface available is also useful for scripting, or for piping in context/prompts from other commands.
A bit more on some of Plandex\u2019s key features:
- Plandex has a built-in diff review sandbox that helps you get the benefits of AI without leaving behind a mess in your project. By default, all changes accumulate in the sandbox until you approve them. The sandbox is version-controlled. You can rewind it to any previous point, and you can also create branches to try out alternative approaches.
- It offers a \u2018full auto mode\u2019 that can complete large tasks autonomously end-to-end, including high level planning, context loading, detailed planning, implementation, command execution (for dependencies, builds, tests, etc.), and debugging.
- The autonomy level is highly configurable. You can move up and down the ladder of autonomy depending on the task, your comfort level, and how you weigh cost optimization vs. effort and results. The default autonomy level is \u2018semi-auto\u2019, which loads context automatically, but requires your approval before applying changes to your project and executing commands.
- Models and model settings are also very configurable. There are built-in models and model packs for different use cases, including `daily-driver` (the default pack), `reasoning` (to use Claude 3.7 thinking in place of plain 3.7), `strong` (for the toughest tasks where extra cost and latency are acceptable), `cheap` (for smaller tasks), and `oss` (for an all open source pack, which currently leans strongly on DeepSeek R1/V3), and `gemini-exp` for a pack fronted by Gemini 2.5 pro. You can also add custom models and model packs, and customize model settings like temperature or top-p. All model changes are version controlled, so you can use branches to try out the same task with different models.
- It can be easily self-hosted, including a \u2018local mode\u2019 for a very fast local single-user setup with Docker. Self-hosting requires OpenAI and OpenRouter.ai accounts/api keys.
- Cloud hosting is also available for added convenience with a couple of subscription tiers: an \u2018Integrated Models\u2019 mode that requires no other accounts or API keys and allows you to manage billing/budgeting/spending alerts and track usage centrally, and a \u2018BYO API Key\u2019 mode that allows you to use your own OpenAI/OpenRouter accounts.
I\u2019m having a blast building Plandex. The more I improve it, the faster I can add features to Plandex itself, which is a great motivator! I have a long list of feature ideas that I think can make a big impact, and I\u2019m also working on building a sustainable and successful company around Plandex so that both the product side and the open source project side can continue to grow and support each other\u2019s growth.
On that note, if you\u2019re interested in working on Plandex in some capacity, either as a contributor or for pay, please either jump into the repo (here\u2019s a development guide - https://docs.plandex.ai/development) or get in touch (via email - dane@plandex.ai)\u2014while I\u2019m not hiring at the moment, I hope to start before too long and would love to connect with talented folks who might be interested in joining me in the future.
I\u2019d also love to get more HNers in the Plandex Discord (https://discord.gg/plandex-ai). Please join and say hi!
And of course I\u2019d love to hear your feedback, whether positive or negative. Thanks so much!"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Plandex v2 \u2013 OSS AI coding agent w/ diff sandbox, full auto, 2M context"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://plandex.ai"}},"_tags":["story","author_danenania","story_43569813","show_hn"],"author":"danenania","children":[43569822],"created_at":"2025-04-03T14:01:02Z","created_at_i":1743688862,"num_comments":0,"objectID":"43569813","points":4,"story_id":43569813,"story_text":"Hey HN! I\u2019m Dane, the creator of Plandex (https://github.com/plandex-ai/plandex), an open source AI coding agent focused especially on tackling large tasks in real world software projects.
You can watch a 2 minute demo of Plandex in action here: https://www.youtube.com/watch?v=SFSu2vNmlLk
And here\u2019s more of a tutorial style demo showing how Plandex can automatically debug a browser app: https://www.youtube.com/watch?v=VCegxOCAPq0
I launched Plandex v1 here on HN a little less than a year ago (https://news.ycombinator.com/item?id=39918500). It got a positive reception overall (plus some skepticism, which is fair enough), and a lot of early users and contributors. HNers helped to push it to the top repo on GitHub trending, where it remained for a few days, racking up over 8k stars.
Now I\u2019m launching a major update, Plandex v2, which is the result of 8 months of heads down work, and is in effect a whole new project/product.
In short, Plandex is now a top-tier coding agent with fully autonomous capabilities.
It combines models from Anthropic, OpenAI, and Google to achieve better results, more reliable agent behavior, better cost efficiency, and better performance than is possible by using only a single provider\u2019s models.
I believe it is now one of the best tools available for working on large tasks in real world codebases with AI. It has an effective context window of 2M tokens, and can index projects of 20M tokens and beyond using tree-sitter project maps (30+ languages are supported). It can effectively find relevant context in massive million-line projects like SQLite, Redis, and Git.
There are also some major UX improvements. Plandex v1 offered just a git-like CLI interface. You\u2019d send prompts via the `plandex tell` command, either passing a string, a prompt file path, or popping up vim/nano.
While that CLI interface still exists, there\u2019s now a much more intuitive chat-like REPL experience that gives you fuzzy autocomplete for commands and file paths. This is the new default/recommended way to use Plandex, though having the CLI interface available is also useful for scripting, or for piping in context/prompts from other commands.
A bit more on some of Plandex\u2019s key features:
- Plandex has a built-in diff review sandbox that helps you get the benefits of AI without leaving behind a mess in your project. By default, all changes accumulate in the sandbox until you approve them. The sandbox is version-controlled. You can rewind it to any previous point, and you can also create branches to try out alternative approaches.
- It offers a \u2018full auto mode\u2019 that can complete large tasks autonomously end-to-end, including high level planning, context loading, detailed planning, implementation, command execution (for dependencies, builds, tests, etc.), and debugging.
- The autonomy level is highly configurable. You can move up and down the ladder of autonomy depending on the task, your comfort level, and how you weigh cost optimization vs. effort and results. The default autonomy level is \u2018semi-auto\u2019, which loads context automatically, but requires your approval before applying changes to your project and executing commands.
- Models and model settings are also very configurable. There are built-in models and model packs for different use cases, including `daily-driver` (the default pack), `reasoning` (to use Claude 3.7 thinking in place of plain 3.7), `strong` (for the toughest tasks where extra cost and latency are acceptable), `cheap` (for smaller tasks), and `oss` (for an all open source pack, which currently leans strongly on DeepSeek R1/V3), and `gemini-exp` for a pack fronted by Gemini 2.5 pro. You can also add custom models and model packs, and customize model settings like temperature or top-p. All model changes are version controlled, so you can use branches to try out the same task with different models.
- It can be easily self-hosted, including a \u2018local mode\u2019 for a very fast local single-user setup with Docker. Self-hosting requires OpenAI and OpenRouter.ai accounts/api keys.
- Cloud hosting is also available for added convenience with a couple of subscription tiers: an \u2018Integrated Models\u2019 mode that requires no other accounts or API keys and allows you to manage billing/budgeting/spending alerts and track usage centrally, and a \u2018BYO API Key\u2019 mode that allows you to use your own OpenAI/OpenRouter accounts.
I\u2019m having a blast building Plandex. The more I improve it, the faster I can add features to Plandex itself, which is a great motivator! I have a long list of feature ideas that I think can make a big impact, and I\u2019m also working on building a sustainable and successful company around Plandex so that both the product side and the open source project side can continue to grow and support each other\u2019s growth.
On that note, if you\u2019re interested in working on Plandex in some capacity, either as a contributor or for pay, please either jump into the repo (here\u2019s a development guide - https://docs.plandex.ai/development) or get in touch (via email - dane@plandex.ai)\u2014while I\u2019m not hiring at the moment, I hope to start before too long and would love to connect with talented folks who might be interested in joining me in the future.
I\u2019d also love to get more HNers in the Plandex Discord (https://discord.gg/plandex-ai). Please join and say hi!
And of course I\u2019d love to hear your feedback, whether positive or negative. Thanks so much!","title":"Show HN: Plandex v2 \u2013 OSS AI coding agent w/ diff sandbox, full auto, 2M context","updated_at":"2025-04-06T22:59:17Z","url":"https://plandex.ai"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Arhan-w"},"story_text":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"GoatCode is a single 85 MB binary that runs in your terminal. It brings 180+ LLM providers (or your existing Claude/ChatGPT/Gemini/Copilot subscriptions via OAuth), and when your quota dies mid-session it automatically retries then walks your fallback chain live \u2014 switches provider mid-turn, says so, and finishes the answer. No crash, no lost context.
What\u2019s new in v2.1.10:\n- Prompt caching on Anthropic (up to 90% cheaper long sessions; savings shown in /cost)\n- Inline unified diffs after every write/edit\n- /rewind \u2014 jump transcript + files back to any turn\n- Parallel subagents (up to 3 task calls fan out concurrently)\n- /search + goat search \u2014 full-text across every saved session\n- @file Tab completion, /model sonnet aliases\n- True context meter (real API tokens vs model window)\n- GOAT MODE (bypass) that actually works: mid-turn switch, prompt resolves YES\n- Noir theme: deep slate + black-green surfaces
Install: npm install -g goatcode-cli (or grab a static binary from the releases page)
Repo: https://github.com/Arhan-w/GoatCode\nLanding: https://goatcode.vercel.app"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"GoatCode \u2013 open-source terminal AI agent with provider failover"}},"_tags":["story","author_Arhan-w","story_49670455","ask_hn"],"author":"Arhan-w","children":[49670922,49683484],"created_at":"2026-09-12T09:02:29Z","created_at_i":1789203749,"num_comments":2,"objectID":"49670455","points":3,"story_id":49670455,"story_text":"GoatCode is a single 85 MB binary that runs in your terminal. It brings 180+ LLM providers (or your existing Claude/ChatGPT/Gemini/Copilot subscriptions via OAuth), and when your quota dies mid-session it automatically retries then walks your fallback chain live \u2014 switches provider mid-turn, says so, and finishes the answer. No crash, no lost context.
What\u2019s new in v2.1.10:\n- Prompt caching on Anthropic (up to 90% cheaper long sessions; savings shown in /cost)\n- Inline unified diffs after every write/edit\n- /rewind \u2014 jump transcript + files back to any turn\n- Parallel subagents (up to 3 task calls fan out concurrently)\n- /search + goat search \u2014 full-text across every saved session\n- @file Tab completion, /model sonnet aliases\n- True context meter (real API tokens vs model window)\n- GOAT MODE (bypass) that actually works: mid-turn switch, prompt resolves YES\n- Noir theme: deep slate + black-green surfaces
Install: npm install -g goatcode-cli (or grab a static binary from the releases page)
Repo: https://github.com/Arhan-w/GoatCode\nLanding: https://goatcode.vercel.app","title":"GoatCode \u2013 open-source terminal AI agent with provider failover","updated_at":"2026-09-15T22:56:41Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ekalabya"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I've been working on Ghost Widget \u2013 an AI memory assistant that captures your screen activity and lets you (or AI agents) search through everything you've ever seen.\nThe problem I'm solving: You spend 20 minutes looking for that article you read last week, or that code snippet you wrote yesterday, or remembering what your manager said in that Zoom call. Your computer has a terrible memory, even though it literally has perfect storage.\nHow it works: Runs in background, captures screen metadata/activity, builds a searchable knowledge base. You can ask "what was that Python library I looked at last Tuesday?" or let Claude/ChatGPT query it directly for context-aware assistance.\nEarly traction: 10 people signed up for demos in the first week with basically no marketing and a landing page I'm embarrassed to show you.\nQuestions for HN:
Privacy: This is obviously the biggest concern. Would local-only storage + open-source code make you comfortable? Or is screen recording a hard no regardless?\nUse cases: Beyond "finding stuff I've seen," what would make this genuinely indispensable for you?\nPricing: Would you pay $20/month for this if it saved you an hour daily? Or does it need to be freemium?\nCompetition: Rewind exists but is Mac-only and just got acquired/pivoted. Microsoft Recall was a privacy nightmare. What would differentiate a new player here?
For context: I'm a 9th grader from India, top 1% on Codeforces, built an AI IDE in 2022 before Cursor was a thing. I ship fast and want to make sure I'm building something people actually want before going deeper.
A very bootstrapped landing page: https://www.ghostwidget.com/"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: Would you use an AI that remembers everything you've done on your PC?"}},"_tags":["story","author_ekalabya","story_45668109","ask_hn"],"author":"ekalabya","children":[45668325,45669042,45670212,45672607],"created_at":"2025-10-22T12:35:53Z","created_at_i":1761136553,"num_comments":4,"objectID":"45668109","points":2,"story_id":45668109,"story_text":"I've been working on Ghost Widget \u2013 an AI memory assistant that captures your screen activity and lets you (or AI agents) search through everything you've ever seen.\nThe problem I'm solving: You spend 20 minutes looking for that article you read last week, or that code snippet you wrote yesterday, or remembering what your manager said in that Zoom call. Your computer has a terrible memory, even though it literally has perfect storage.\nHow it works: Runs in background, captures screen metadata/activity, builds a searchable knowledge base. You can ask "what was that Python library I looked at last Tuesday?" or let Claude/ChatGPT query it directly for context-aware assistance.\nEarly traction: 10 people signed up for demos in the first week with basically no marketing and a landing page I'm embarrassed to show you.\nQuestions for HN:
Privacy: This is obviously the biggest concern. Would local-only storage + open-source code make you comfortable? Or is screen recording a hard no regardless?\nUse cases: Beyond "finding stuff I've seen," what would make this genuinely indispensable for you?\nPricing: Would you pay $20/month for this if it saved you an hour daily? Or does it need to be freemium?\nCompetition: Rewind exists but is Mac-only and just got acquired/pivoted. Microsoft Recall was a privacy nightmare. What would differentiate a new player here?
For context: I'm a 9th grader from India, top 1% on Codeforces, built an AI IDE in 2022 before Cursor was a thing. I ship fast and want to make sure I'm building something people actually want before going deeper.
A very bootstrapped landing page: https://www.ghostwidget.com/","title":"Ask HN: Would you use an AI that remembers everything you've done on your PC?","updated_at":"2026-03-05T22:52:39Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ajabhish"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"Demo: https://youtu.be/W4uBOTf9KtY
Today we\u2019re rolling out beta access to our AI agent that talks to Google Slides.\n- Ask a question \u2192 it jumps to the right slide\n- Need a diagram? \u2192 \u201cShow slide 3\u201d\n- Want to rewind? \u2192 \u201cGo back two slides\u201d
This can be a game changer for teaching new concepts, employee training (onboarding, explaining playbooks etc) ...
Our agent now has capability to display and navigate Google Slides based on conversation flow, creating truly interactive learning experiences.
Setup in 3 steps:\n 1\u20e3 Get the embed URL from Google Slides (File \u2192 Share \u2192 Publish to web \u2192 Embed)\n 2\u20e3 Enable the tool in scenario settings and paste the URL\n 3\u20e3 Add slide instructions to your AI prompt (e.g., \u201cShow slide 3 for the architecture diagram\u201d)
Create your own: https://app.toughtongueai.com/advanced-scenario-create/
Try the one I used: https://app.toughtongueai.com/run/paris-tough-tongue-ai-pitc...
Or the one that teaches: https://app.toughtongueai.com/library/product-design-intervi..."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: Transform static presentations \u2192 dynamic AI-guided experiences"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://app.toughtongueai.com/library/product-design-interview-tips-5-strategies-688bb9ca8021d1a72a3d25fa/"}},"_tags":["story","author_ajabhish","story_44753245","show_hn"],"author":"ajabhish","children":[44753257],"created_at":"2025-08-01T05:19:30Z","created_at_i":1754025570,"num_comments":0,"objectID":"44753245","points":2,"story_id":44753245,"story_text":"Demo: https://youtu.be/W4uBOTf9KtY
Today we\u2019re rolling out beta access to our AI agent that talks to Google Slides.\n- Ask a question \u2192 it jumps to the right slide\n- Need a diagram? \u2192 \u201cShow slide 3\u201d\n- Want to rewind? \u2192 \u201cGo back two slides\u201d
This can be a game changer for teaching new concepts, employee training (onboarding, explaining playbooks etc) ...
Our agent now has capability to display and navigate Google Slides based on conversation flow, creating truly interactive learning experiences.
Setup in 3 steps:\n 1\u20e3 Get the embed URL from Google Slides (File \u2192 Share \u2192 Publish to web \u2192 Embed)\n 2\u20e3 Enable the tool in scenario settings and paste the URL\n 3\u20e3 Add slide instructions to your AI prompt (e.g., \u201cShow slide 3 for the architecture diagram\u201d)
Create your own: https://app.toughtongueai.com/advanced-scenario-create/
Try the one I used: https://app.toughtongueai.com/run/paris-tough-tongue-ai-pitc...
Or the one that teaches: https://app.toughtongueai.com/library/product-design-intervi...","title":"Show HN: Transform static presentations \u2192 dynamic AI-guided experiences","updated_at":"2025-08-01T05:23:16Z","url":"https://app.toughtongueai.com/library/product-design-interview-tips-5-strategies-688bb9ca8021d1a72a3d25fa/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"celene-II_thro"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"I've been watching things progress & my mind keeps wondering about the potential for using AI to exonerate the wrongly convicted. Specifically in cases of things like 'Murder by Proxy', where there's a much higher probability that those pulling the strings will have sophisticated deniability tactics in place from which the 'patsy' has no hope of escape.
I know how since as far back as the Gulf war we've had the ability to fly drones 24/7 at altitudes far out of sight to provide the ability to rewind & playback anyone's movements along with anyone/anything they'd have encountered or even had a passing glance at.
My thinking is that if we can feed this to a sufficiently trained AI we could expose deeply-entrenched underground MBP networks which have likely flourished & evolved at the hands of organized crime syndicates for god-only-knows how long now.
I know there's always the concern for civil-rights-type abuse of such machinery, but assuming we're able to protect ourselves from the grey-goo problem, it can be assumed this issue is moot as well."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Ask HN: Anyone Use AI in a Forensic Setting?"}},"_tags":["story","author_celene-II_thro","story_34037533","ask_hn"],"author":"celene-II_thro","children":[34037564],"created_at":"2022-12-18T13:42:14Z","created_at_i":1671370934,"num_comments":1,"objectID":"34037533","points":1,"story_id":34037533,"story_text":"I've been watching things progress & my mind keeps wondering about the potential for using AI to exonerate the wrongly convicted. Specifically in cases of things like 'Murder by Proxy', where there's a much higher probability that those pulling the strings will have sophisticated deniability tactics in place from which the 'patsy' has no hope of escape.
I know how since as far back as the Gulf war we've had the ability to fly drones 24/7 at altitudes far out of sight to provide the ability to rewind & playback anyone's movements along with anyone/anything they'd have encountered or even had a passing glance at.
My thinking is that if we can feed this to a sufficiently trained AI we could expose deeply-entrenched underground MBP networks which have likely flourished & evolved at the hands of organized crime syndicates for god-only-knows how long now.
I know there's always the concern for civil-rights-type abuse of such machinery, but assuming we're able to protect ourselves from the grey-goo problem, it can be assumed this issue is moot as well.","title":"Ask HN: Anyone Use AI in a Forensic Setting?","updated_at":"2024-09-20T12:50:06Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"echohive42"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"A Byzantine mosaic you can rewind"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://www.echohive.ai/experiments/hippodrome"}},"_tags":["story","author_echohive42","story_49859120"],"author":"echohive42","created_at":"2026-09-26T18:23:50Z","created_at_i":1790447030,"num_comments":0,"objectID":"49859120","points":1,"story_id":49859120,"title":"A Byzantine mosaic you can rewind","updated_at":"2026-09-26T18:27:29Z","url":"https://www.echohive.ai/experiments/hippodrome"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"KarinneLima"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["rewind"],"value":"Recover Old Slack Links with Paperbot Rewind"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["rewind","ai"],"value":"http://help.paperbot.ai/browsing-and-searching/new-restoring-links-history-with-rewind?utm_source=HN&utm_medium=rewind"}},"_tags":["story","author_KarinneLima","story_14469222"],"author":"KarinneLima","created_at":"2017-06-02T11:58:07Z","created_at_i":1496404687,"num_comments":0,"objectID":"14469222","points":1,"story_id":14469222,"title":"Recover Old Slack Links with Paperbot Rewind","updated_at":"2024-09-20T00:54:57Z","url":"http://help.paperbot.ai/browsing-and-searching/new-restoring-links-history-with-rewind?utm_source=HN&utm_medium=rewind"}],"hitsPerPage":50,"nbHits":152,"nbPages":4,"page":0,"params":"query=Rewind+AI&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":11,"processingTimingsMS":{"_request":{"roundTrip":24},"afterFetch":{"format":{"highlighting":2,"total":3},"merge":{"mergeLoop":{"prepareNextHit":3,"total":3},"total":3},"total":3},"fetch":{"query":2,"scanning":4,"total":7},"total":11},"query":"Rewind AI","serverTimeMS":15}