I am around 30 and feel burnt out from my tech career. I do not feel any interest or happiness towards building things right now. I chose this career mostly due to monetary reasons and pursued undergrad and graduate degrees in CS in Europe. After which, I worked as a product manager for a while in the networking/telco sector in internal enterprise systems.
I am really at a loss for what to do now or how to spend time. I was always interested in doing my own thing, but would not call myself a good engineer or as having deep skills in a particular domain. I am more of a jack of all trades.
I dread just getting the next job and continuing the corporate grind, but I really worry about what to do in the near future career-wise.\nIf you were in my position, what would you do with the time?"},"title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: 30, burnt out, with 6-10 months of runway. What would you do?"}},"_tags":["story","author_throwaway_30yo","story_49535363","ask_hn"],"author":"throwaway_30yo","children":[49559278,49559475,49559484,49559560,49559695,49559883,49560351,49560398,49560516,49560631,49560735,49561331,49561383,49561632,49561748,49561926,49562442,49562604,49563264,49563946,49566046,49568575,49571783,49579329,49579657,49580075,49580357,49581939],"created_at":"2026-09-02T12:39:09Z","created_at_i":1788352749,"num_comments":26,"objectID":"49535363","points":31,"story_id":49535363,"story_text":"After being let go from a small agency due to financial reasons, I find myself having at least 6-10 months of downtime for the first time in 10 years.
I am around 30 and feel burnt out from my tech career. I do not feel any interest or happiness towards building things right now. I chose this career mostly due to monetary reasons and pursued undergrad and graduate degrees in CS in Europe. After which, I worked as a product manager for a while in the networking/telco sector in internal enterprise systems.
I am really at a loss for what to do now or how to spend time. I was always interested in doing my own thing, but would not call myself a good engineer or as having deep skills in a particular domain. I am more of a jack of all trades.
I dread just getting the next job and continuing the corporate grind, but I really worry about what to do in the near future career-wise.\nIf you were in my position, what would you do with the time?","title":"Ask HN: 30, burnt out, with 6-10 months of runway. What would you do?","updated_at":"2026-09-07T07:42:43Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"theonlinx"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"Hi everyone, I'm 22 years old and I own a start up with my co-founder, we both just graduated college and we built a b2b SaaS that's live with 60 users and growing. We've been live since beginning of June.
I wanted to post on here because we need help, real help and we don't know where to look or who to ask or how to find reliable people. We see teams like Cluely and Whop and obviously they're billion dollar companies but how do people in start ups find a team like that? Not just employees but a team, a family almost. It is just us two and we're totally boot strapped and dialed in and we've been making it work I think really really well but as we keep growing new things keep coming up and new tasks to do, new things to learn as a company grows especially since its our first.
I was just wondering maybe some stories of how other people have found the rest of their team, that doesn't involve moving out to SF lol.
We started on our project October 2025, had the MVP built January/February and some free live users on there, then we officially launched to the pubic in June.
We are the sales guys, customer support, coders, marketers, content creators, managers, just everything right now. And we both just started our full time jobs that we graduated college for, that have nothing to do with our startup, the entire summer we were free to work on everything and that's why we got so far but now we're not going to have as much time.
Our software is working and people are interested in it and using it. There's real potential for this to be something, especially just based on the traction we've gotten with regular organic content on TikTok and one meta ad.
The goal is to be able to fully fund ourselves from the startup, go all in on it where we can devote our full time attention to it and keep growing it and the team.
We're babies at this thing, we're basically still in the womb and we're just trying to absorb as much information and guidance as possible, sorry I chose not to name the software here, just looking for advice on building a real team and I'd love to hear your guys' stories.
Thanks"},"title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: How do we build a team? (bootstrapped B2B startup)"}},"_tags":["story","author_theonlinx","story_49556698","ask_hn"],"author":"theonlinx","children":[49570482],"created_at":"2026-09-03T20:46:09Z","created_at_i":1788468369,"num_comments":1,"objectID":"49556698","points":5,"story_id":49556698,"story_text":"Hi everyone, I'm 22 years old and I own a start up with my co-founder, we both just graduated college and we built a b2b SaaS that's live with 60 users and growing. We've been live since beginning of June.
I wanted to post on here because we need help, real help and we don't know where to look or who to ask or how to find reliable people. We see teams like Cluely and Whop and obviously they're billion dollar companies but how do people in start ups find a team like that? Not just employees but a team, a family almost. It is just us two and we're totally boot strapped and dialed in and we've been making it work I think really really well but as we keep growing new things keep coming up and new tasks to do, new things to learn as a company grows especially since its our first.
I was just wondering maybe some stories of how other people have found the rest of their team, that doesn't involve moving out to SF lol.
We started on our project October 2025, had the MVP built January/February and some free live users on there, then we officially launched to the pubic in June.
We are the sales guys, customer support, coders, marketers, content creators, managers, just everything right now. And we both just started our full time jobs that we graduated college for, that have nothing to do with our startup, the entire summer we were free to work on everything and that's why we got so far but now we're not going to have as much time.
Our software is working and people are interested in it and using it. There's real potential for this to be something, especially just based on the traction we've gotten with regular organic content on TikTok and one meta ad.
The goal is to be able to fully fund ourselves from the startup, go all in on it where we can devote our full time attention to it and keep growing it and the team.
We're babies at this thing, we're basically still in the womb and we're just trying to absorb as much information and guidance as possible, sorry I chose not to name the software here, just looking for advice on building a real team and I'd love to hear your guys' stories.
Thanks","title":"Ask HN: How do we build a team? (bootstrapped B2B startup)","updated_at":"2026-09-06T21:42:57Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"irscott"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"I make music that's entirely found sound/sample based. It occurred to me that on every social media site (particularly Instagram) I would have thousands of saved posts because of some sound or visual content that I thought I might like to sample someday.
I made Harvester, a local-first Firefox extension for collecting source material for sampling, collage, VJing, editing, and research. Harvester natively preserves media from individual Instagram, Youtube, and Reddit posts as well as visible video and audio from other websites that expose ordinary audio or video. Harvester not only saves the original video to your archive but it also creates a separate audio only file in your chosen format and saves a metadata.json file containing provenance info.
Harvester can scan multiple Instagram Saved collections and create oldest-first queues, allowing you to archive posts gradually in small, deliberately paced batches. Built-in limits and randomized delays help reduce the risks associated with automated activity, though no tool can guarantee that Instagram won\u2019t restrict your account. It also supports carousel posts, grabbing each video and creating the audio only derivative for each video in the carousel.
There's no Harvester account, telemetry, or other wacky stuff built in. Just a basic tool designed to help creators make use of the interesting sights and sounds we encounter online every day.
This is just the first release, currently for Firefox on MacOS and it does require a local companion to work properly. I'd especially appreciate feedback about the installation process and sources where it fails.
The extension is plain JavaScript, and the local Python companion uses yt-dlp and FFmpeg for bounded acquisition and media processing.
https://github.com/sxrvys/harvester"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Harvester, a local first media collection tool for sample based artists"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/sxrvys/harvester"}},"_tags":["story","author_irscott","story_49545211","show_hn"],"author":"irscott","created_at":"2026-09-03T02:20:02Z","created_at_i":1788402002,"num_comments":0,"objectID":"49545211","points":3,"story_id":49545211,"story_text":"I make music that's entirely found sound/sample based. It occurred to me that on every social media site (particularly Instagram) I would have thousands of saved posts because of some sound or visual content that I thought I might like to sample someday.
I made Harvester, a local-first Firefox extension for collecting source material for sampling, collage, VJing, editing, and research. Harvester natively preserves media from individual Instagram, Youtube, and Reddit posts as well as visible video and audio from other websites that expose ordinary audio or video. Harvester not only saves the original video to your archive but it also creates a separate audio only file in your chosen format and saves a metadata.json file containing provenance info.
Harvester can scan multiple Instagram Saved collections and create oldest-first queues, allowing you to archive posts gradually in small, deliberately paced batches. Built-in limits and randomized delays help reduce the risks associated with automated activity, though no tool can guarantee that Instagram won\u2019t restrict your account. It also supports carousel posts, grabbing each video and creating the audio only derivative for each video in the carousel.
There's no Harvester account, telemetry, or other wacky stuff built in. Just a basic tool designed to help creators make use of the interesting sights and sounds we encounter online every day.
This is just the first release, currently for Firefox on MacOS and it does require a local companion to work properly. I'd especially appreciate feedback about the installation process and sources where it fails.
The extension is plain JavaScript, and the local Python companion uses yt-dlp and FFmpeg for bounded acquisition and media processing.
https://github.com/sxrvys/harvester","title":"Show HN: Harvester, a local first media collection tool for sample based artists","updated_at":"2026-09-06T18:09:26Z","url":"https://github.com/sxrvys/harvester"},{"_highlightResult":{"author":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"gradus_ad"},"story_text":{"matchLevel":"none","matchedWords":[],"value":"Model sycophancy is going to become a major issue for individuals and society... Do any benchmarks exist for this? Really needs to become a standard aspect of model quality measures"},"title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: Model sycophancy benchmarks?"}},"_tags":["story","author_gradus_ad","story_49588541","ask_hn"],"author":"gradus_ad","children":[49589023,49594803],"created_at":"2026-09-06T17:07:10Z","created_at_i":1788714430,"num_comments":2,"objectID":"49588541","points":4,"story_id":49588541,"story_text":"Model sycophancy is going to become a major issue for individuals and society... Do any benchmarks exist for this? Really needs to become a standard aspect of model quality measures","title":"Ask HN: Model sycophancy benchmarks?","updated_at":"2026-09-07T06:52:58Z"},{"_highlightResult":{"author":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"gradus_ad"},"title":{"matchLevel":"none","matchedWords":[],"value":"80k-year-old \"arrowheads\" rewrite human history"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.sciencedaily.com/releases/2026/09/260903064233.htm"}},"_tags":["story","author_gradus_ad","story_49582546"],"author":"gradus_ad","created_at":"2026-09-06T01:43:42Z","created_at_i":1788659022,"num_comments":0,"objectID":"49582546","points":4,"story_id":49582546,"title":"80k-year-old \"arrowheads\" rewrite human history","updated_at":"2026-09-06T06:44:55Z","url":"https://www.sciencedaily.com/releases/2026/09/260903064233.htm"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"scrolling"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"LED Sports Ticker \u2013 A high-performance 60fps stadium-style scoreboard"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://play.google.com/store/apps/details?id=com.tickerled.sports&hl=en_US"}},"_tags":["story","author_scrolling","story_49592212"],"author":"scrolling","children":[49592213],"created_at":"2026-09-06T23:59:47Z","created_at_i":1788739187,"num_comments":0,"objectID":"49592212","points":2,"story_id":49592212,"title":"LED Sports Ticker \u2013 A high-performance 60fps stadium-style scoreboard","updated_at":"2026-09-07T01:44:13Z","url":"https://play.google.com/store/apps/details?id=com.tickerled.sports&hl=en_US"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"guyb3"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"Hi HN, Jonathan & Guy here from OneCLI, an agent harness built for teams, giving every employee a secured, sandboxed personal agent.
Here\u2019s what you can do with it:
1. get a sandboxed agent, with all the OneCLI capabilities in place like connect your GitHub account, Gmail, Notion, or Dropbox simply from the chat.
2. deterministic human in the loop approval in the chat itself for things that you need 100% control like sending an email or deleting the Linear ticket.
3. manage team policy in one place, enforced across every agent in the workspace
4. enjoy global connections at the team level, like shared LLM keys or service accounts
Here\u2019s a demo: https://www.youtube.com/watch?v=dlW-44ntpbE
We started working on this by accident, even though our careers were in the security space. We were working on a devtool called ChartDB, an open-source DB tool. When OpenClaw took off back in January, we started using it to orchestrate agents on top of ChartDB. We quickly understood there is a big issue around auth. Agents need credentials to do real work, but to give them those secrets would not be the best idea. They keep them in their memory and also write them down to local files and their sessions as plain text. And we knew that agents can easily be fooled into giving up those API keys/secrets. So we needed some way to control the agent and stop prompt injections from tricking it into using its services for an attacker's benefit.
We created OneCLI that started as a vault for AI Agents built in Rust.
We found out that most of our demand for OneCLI came from autonomous agents like Hermes, OpenClaw and NanoClaw for individuals and teams.
Users looked for useful agents that do things for the person who runs them with two missing parts: 1) managing secrets and permissions. 2) and for teams - multiplayer management.
We decided to pivot and provide the agent itself as a harness for teams, to give each employee an agent. We saw that teams had to deal with setting up their own harness again and again, and basically as we already had the vault as a gateway. We got the idea to provide the missing piece of the agent management out of the box and open source it (Apache-2.0, with a small enterprise exception).
We're open source first - the entire platform, not just a small portion of it like other agents, so companies can actually see the code, evaluate it, and trust it instead of taking our word for it. They run it isolated, in their own environment, fully under their control, at production quality, not a locked black box hosted somewhere else. That means the safety isn't just a promise, it's something they can verify themselves. Combined with real autonomy and least-privilege access, that's what makes it something a company can fully own and trust, not just adopt.
We also approach this from a company perspective rather than an individual one. Our solution manages agents on behalf of each employee, wrapped in deterministic guardrails that company admins configure through centralized policies.
For the agent engine itself we\u2019re using jcode which is the core of the agent-loop. We found out that it improves the experience and makes the agent smarter and faster.
Here\u2019s how it works:
It runs on infra you control. Fully open-source, self-host or cloud in minutes.
The agent never holds a real secret. It gets a placeholder. The real credential is injected at the gateway, per request, after the call is authorized. It never enters the agent's context, memory, or logs.
Enforcement outside the model. Prompts are suggestions. Policies defined by the org admin run at the network layer, outside the agent and the LLM. Block endpoints, rate limit per agent, require approval, scope per employee. The gateway decides. The agent can't bypass it.
Isolated VM per agent. Own memory, own keys, own permissions. Blast radius is one agent.
Speed of the Harness: Rust engine under the agent loop.
Full identity trail. Every agent is bound to an employee. Every call logged with who it acted for and which policy allowed it.
Some things people are doing with the platform include:
- Managing their company life cycle entirely from the sales calls, to the product side automatically open tickets to the engineering teams, that would kick the development agents to deliver and ship to production.
- Operational side, like automatically hygiene the CRM after calls, sourcing leads, book meetings and manage follow ups emails.
- Some of our customers also doing their entire grocery shopping using those agents and send them to take care of their chores like ordering things online.
About the team: Both founders come from cybersecurity backgrounds. Jonathan spent years at Axis Security building zero trust network access. The core idea is that you never trust the client. You decide exactly what a person can reach, and you enforce it outside of them, at the network layer, so it doesn't matter what the client tries to do. That's how every serious company gives access to humans today. Guy was the 1st employee in Argon security doing AppSec.
We would love to hear your thoughts on the move, happy to get issues open to improve and get your agent to be powerful and secure - designed for teams, not just individuals."},"title":{"matchLevel":"none","matchedWords":[],"value":"Launch HN: OneCLI (YC S26) \u2013 OSS sandboxed agent harness for teams"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/onecli/onecli"}},"_tags":["story","author_guyb3","story_49363710","launch_hn"],"author":"guyb3","children":[49364385,49364395,49364991,49365663,49366194,49366268,49367087,49369075,49369922,49370188,49370551,49370999,49371194,49384329,49385790,49390885,49394663,49406200,49412191,49425460,49430925,49432200,49487702,49507688],"created_at":"2026-08-19T16:29:02Z","created_at_i":1787156942,"num_comments":37,"objectID":"49363710","points":88,"story_id":49363710,"story_text":"Hi HN, Jonathan & Guy here from OneCLI, an agent harness built for teams, giving every employee a secured, sandboxed personal agent.
Here\u2019s what you can do with it:
1. get a sandboxed agent, with all the OneCLI capabilities in place like connect your GitHub account, Gmail, Notion, or Dropbox simply from the chat.
2. deterministic human in the loop approval in the chat itself for things that you need 100% control like sending an email or deleting the Linear ticket.
3. manage team policy in one place, enforced across every agent in the workspace
4. enjoy global connections at the team level, like shared LLM keys or service accounts
Here\u2019s a demo: https://www.youtube.com/watch?v=dlW-44ntpbE
We started working on this by accident, even though our careers were in the security space. We were working on a devtool called ChartDB, an open-source DB tool. When OpenClaw took off back in January, we started using it to orchestrate agents on top of ChartDB. We quickly understood there is a big issue around auth. Agents need credentials to do real work, but to give them those secrets would not be the best idea. They keep them in their memory and also write them down to local files and their sessions as plain text. And we knew that agents can easily be fooled into giving up those API keys/secrets. So we needed some way to control the agent and stop prompt injections from tricking it into using its services for an attacker's benefit.
We created OneCLI that started as a vault for AI Agents built in Rust.
We found out that most of our demand for OneCLI came from autonomous agents like Hermes, OpenClaw and NanoClaw for individuals and teams.
Users looked for useful agents that do things for the person who runs them with two missing parts: 1) managing secrets and permissions. 2) and for teams - multiplayer management.
We decided to pivot and provide the agent itself as a harness for teams, to give each employee an agent. We saw that teams had to deal with setting up their own harness again and again, and basically as we already had the vault as a gateway. We got the idea to provide the missing piece of the agent management out of the box and open source it (Apache-2.0, with a small enterprise exception).
We're open source first - the entire platform, not just a small portion of it like other agents, so companies can actually see the code, evaluate it, and trust it instead of taking our word for it. They run it isolated, in their own environment, fully under their control, at production quality, not a locked black box hosted somewhere else. That means the safety isn't just a promise, it's something they can verify themselves. Combined with real autonomy and least-privilege access, that's what makes it something a company can fully own and trust, not just adopt.
We also approach this from a company perspective rather than an individual one. Our solution manages agents on behalf of each employee, wrapped in deterministic guardrails that company admins configure through centralized policies.
For the agent engine itself we\u2019re using jcode which is the core of the agent-loop. We found out that it improves the experience and makes the agent smarter and faster.
Here\u2019s how it works:
It runs on infra you control. Fully open-source, self-host or cloud in minutes.
The agent never holds a real secret. It gets a placeholder. The real credential is injected at the gateway, per request, after the call is authorized. It never enters the agent's context, memory, or logs.
Enforcement outside the model. Prompts are suggestions. Policies defined by the org admin run at the network layer, outside the agent and the LLM. Block endpoints, rate limit per agent, require approval, scope per employee. The gateway decides. The agent can't bypass it.
Isolated VM per agent. Own memory, own keys, own permissions. Blast radius is one agent.
Speed of the Harness: Rust engine under the agent loop.
Full identity trail. Every agent is bound to an employee. Every call logged with who it acted for and which policy allowed it.
Some things people are doing with the platform include:
- Managing their company life cycle entirely from the sales calls, to the product side automatically open tickets to the engineering teams, that would kick the development agents to deliver and ship to production.
- Operational side, like automatically hygiene the CRM after calls, sourcing leads, book meetings and manage follow ups emails.
- Some of our customers also doing their entire grocery shopping using those agents and send them to take care of their chores like ordering things online.
About the team: Both founders come from cybersecurity backgrounds. Jonathan spent years at Axis Security building zero trust network access. The core idea is that you never trust the client. You decide exactly what a person can reach, and you enforce it outside of them, at the network layer, so it doesn't matter what the client tries to do. That's how every serious company gives access to humans today. Guy was the 1st employee in Argon security doing AppSec.
We would love to hear your thoughts on the move, happy to get issues open to improve and get your agent to be powerful and secure - designed for teams, not just individuals.","title":"Launch HN: OneCLI (YC S26) \u2013 OSS sandboxed agent harness for teams","updated_at":"2026-09-06T00:18:10Z","url":"https://github.com/onecli/onecli"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"alanfuNZ"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["gradium"],"value":"I've been giving coding agents more autonomy lately, letting them run shell commands unattended for longer stretches, and I don't have a good answer for how people actually gate that beyond "run it in a container and hope." A container limits blast radius but doesn't stop the agent from reading a secret and then making an outbound call in the same session, or force-pushing to a branch it shouldn't touch, or just doing something irreversible while nobody's watching. Curious what people are actually doing: allowlists of commands, human-in-the-loop approval for anything destructive, something built into the agent framework itself, or just accepting the risk because the alternative is too slow? Specifically interested in what happens when the approval step itself fails or times out, does your setup default to allow or deny?"},"title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: How do you gate an autonomous coding agent's shell access?"}},"_tags":["story","author_alanfuNZ","story_49556858","ask_hn"],"author":"alanfuNZ","children":[49559118,49561621,49572246,49589759],"created_at":"2026-09-03T20:56:36Z","created_at_i":1788468996,"num_comments":3,"objectID":"49556858","points":2,"story_id":49556858,"story_text":"I've been giving coding agents more autonomy lately, letting them run shell commands unattended for longer stretches, and I don't have a good answer for how people actually gate that beyond "run it in a container and hope." A container limits blast radius but doesn't stop the agent from reading a secret and then making an outbound call in the same session, or force-pushing to a branch it shouldn't touch, or just doing something irreversible while nobody's watching. Curious what people are actually doing: allowlists of commands, human-in-the-loop approval for anything destructive, something built into the agent framework itself, or just accepting the risk because the alternative is too slow? Specifically interested in what happens when the approval step itself fails or times out, does your setup default to allow or deny?","title":"Ask HN: How do you gate an autonomous coding agent's shell access?","updated_at":"2026-09-06T19:10:56Z"}],"hitsPerPage":50,"nbHits":30,"nbPages":1,"page":0,"params":"query=Gradium&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":6,"processingTimingsMS":{"_request":{"roundTrip":19},"afterFetch":{"format":{"highlighting":2,"total":2},"merge":{"mergeLoop":{"prepareNextHit":1,"total":1},"total":1},"total":1},"fetch":{"query":3,"total":4},"total":7},"query":"Gradium","serverTimeMS":9}