Our open source repo is at https://github.com/inconvoai/inconvo (`npx inconvo@latest dev` to run locally), we also have a cloud version (https://app.inconvo.ai) :)
Here\u2019s a quick demo of Inconvo: https://youtu.be/sbDTAbVG-WQ
We built Inconvo to solve a specific problem: building a chat-with-data demo is easy, but shipping to production for real customers is much harder.
We previously did a Launch HN (https://news.ycombinator.com/item?id=44984096). Since then, we\u2019ve open-sourced the core: the entire data agent, semantic layer, and database connectors \u2014 under Apache 2.0, so developers can run it locally end-to-end, inspect the code, and contribute.
For a chat-with-data build, most teams start by showing an LLM the schema and letting it generate SQL. LLMs are great at SQL, so it's easy to get a demo working quickly, but going to production is where it breaks: you need to hide certain tables/columns, handle role based access to data and give the model a better understanding of your business logic and how it maps to the data. Some teams use prompts to attempt to solve these issues, but prompts are just suggestions to a probabilistic system that can emit arbitrary queries. For production, a set of guardrails and checks are required to make sure the system is behaving to your specification at all times.
Inconvo flips the approach: constrain the agent up front instead of trying to rein it in after. You define a semantic layer as the contract (approved tables/columns, metrics, filters/enums, join paths, tenancy rules). At runtime, the model outputs a structured query plan in the form of an IL (intermediate language) not SQL. We validate the IL deterministically against the semantic layer contract, and only then compile and execute it, so tenant scoping and field restrictions etc. are enforced by deterministic code on every request.
Here\u2019s a simplified example of the structured query plan (IL) the model emits:
```\n{\n "table": "reviews",\n "where": [{ "organisation_id": { "equals": 1 } }],\n "operation": "groupBy",\n "operationParameters": {\n "joins": [{ "table": "products", "joinPath": "reviews.product", "joinType": "inner" }],\n "groupBy": ["products.title"],\n "avg": { "columns": ["reviews.rating"] },\n "orderBy": { "function": "avg", "column": "reviews.rating", "direction": "desc" },\n "limit": 1\n }\n}\n```
In a nutshell, Inconvo data agents have a smaller set of safe moves they combine creatively, instead of an infinite surface area to validate after the fact.
There is a real tradeoff here: letting the model generate SQL is maximally flexible, and works well for systems with technical humans-in-the-loop who understand the data (e.g internal BI). Inconvo is more constrained by design so that behavior is enforceable for customer-facing use cases.
Because Inconvo is designed to be embedded in customer-facing products, we\u2019re focused on the developer experience. Applications call Inconvo agents from their backend via our API or TypeScript SDK, and responses come back as typed, structured JSON (text, table, or chart as Vega-Lite spec) so they can be rendered directly in the UI. In multi-agent setups e.g. an application wide agent, it also fits nicely as a specialist chat-with-data sub-agent for the orchestrator.
We\u2019re really happy to show this to you all. Thanks for reading. Please let us know your thoughts and questions in the comments."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Show HN: Inconvo \u2013 open-source chat-with-data agent that doesn't generate SQL"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/inconvoai/inconvo"}},"_tags":["story","author_ogham","story_47078279","show_hn"],"author":"ogham","children":[47086037,47223239],"created_at":"2026-02-19T19:51:59Z","created_at_i":1771530719,"num_comments":1,"objectID":"47078279","points":3,"story_id":47078279,"story_text":"Hey HN, we\u2019re building Inconvo (https://inconvo.com) \u2014 an open-source tool to help developers build chat-with-data agents for their products.
Our open source repo is at https://github.com/inconvoai/inconvo (`npx inconvo@latest dev` to run locally), we also have a cloud version (https://app.inconvo.ai) :)
Here\u2019s a quick demo of Inconvo: https://youtu.be/sbDTAbVG-WQ
We built Inconvo to solve a specific problem: building a chat-with-data demo is easy, but shipping to production for real customers is much harder.
We previously did a Launch HN (https://news.ycombinator.com/item?id=44984096). Since then, we\u2019ve open-sourced the core: the entire data agent, semantic layer, and database connectors \u2014 under Apache 2.0, so developers can run it locally end-to-end, inspect the code, and contribute.
For a chat-with-data build, most teams start by showing an LLM the schema and letting it generate SQL. LLMs are great at SQL, so it's easy to get a demo working quickly, but going to production is where it breaks: you need to hide certain tables/columns, handle role based access to data and give the model a better understanding of your business logic and how it maps to the data. Some teams use prompts to attempt to solve these issues, but prompts are just suggestions to a probabilistic system that can emit arbitrary queries. For production, a set of guardrails and checks are required to make sure the system is behaving to your specification at all times.
Inconvo flips the approach: constrain the agent up front instead of trying to rein it in after. You define a semantic layer as the contract (approved tables/columns, metrics, filters/enums, join paths, tenancy rules). At runtime, the model outputs a structured query plan in the form of an IL (intermediate language) not SQL. We validate the IL deterministically against the semantic layer contract, and only then compile and execute it, so tenant scoping and field restrictions etc. are enforced by deterministic code on every request.
Here\u2019s a simplified example of the structured query plan (IL) the model emits:
```\n{\n "table": "reviews",\n "where": [{ "organisation_id": { "equals": 1 } }],\n "operation": "groupBy",\n "operationParameters": {\n "joins": [{ "table": "products", "joinPath": "reviews.product", "joinType": "inner" }],\n "groupBy": ["products.title"],\n "avg": { "columns": ["reviews.rating"] },\n "orderBy": { "function": "avg", "column": "reviews.rating", "direction": "desc" },\n "limit": 1\n }\n}\n```
In a nutshell, Inconvo data agents have a smaller set of safe moves they combine creatively, instead of an infinite surface area to validate after the fact.
There is a real tradeoff here: letting the model generate SQL is maximally flexible, and works well for systems with technical humans-in-the-loop who understand the data (e.g internal BI). Inconvo is more constrained by design so that behavior is enforceable for customer-facing use cases.
Because Inconvo is designed to be embedded in customer-facing products, we\u2019re focused on the developer experience. Applications call Inconvo agents from their backend via our API or TypeScript SDK, and responses come back as typed, structured JSON (text, table, or chart as Vega-Lite spec) so they can be rendered directly in the UI. In multi-agent setups e.g. an application wide agent, it also fits nicely as a specialist chat-with-data sub-agent for the orchestrator.
We\u2019re really happy to show this to you all. Thanks for reading. Please let us know your thoughts and questions in the comments.","title":"Show HN: Inconvo \u2013 open-source chat-with-data agent that doesn't generate SQL","updated_at":"2026-03-05T23:35:57Z","url":"https://github.com/inconvoai/inconvo"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"shcheklein"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"OpenAI's Data Agent and the S3 Gap"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://datachain.ai/blog/openai-data-agent-s3-gap"}},"_tags":["story","author_shcheklein","story_48049439"],"author":"shcheklein","created_at":"2026-05-07T13:49:27Z","created_at_i":1778161767,"num_comments":0,"objectID":"48049439","points":3,"story_id":48049439,"title":"OpenAI's Data Agent and the S3 Gap","updated_at":"2026-05-07T14:23:15Z","url":"https://datachain.ai/blog/openai-data-agent-s3-gap"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"anophelon"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Show HN: I made a data agent for hypothesis testing"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://github.com/antonguzun/data-agent"}},"_tags":["story","author_anophelon","story_41983706","show_hn"],"author":"anophelon","created_at":"2024-10-29T13:36:34Z","created_at_i":1730208994,"num_comments":0,"objectID":"41983706","points":3,"story_id":41983706,"title":"Show HN: I made a data agent for hypothesis testing","updated_at":"2024-10-29T13:40:47Z","url":"https://github.com/antonguzun/data-agent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"kimoz"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"\"Lambda: A Large Model Based Data Agent\""},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/Stephen-SMJ/LAMBDA"}},"_tags":["story","author_kimoz","story_41103530"],"author":"kimoz","created_at":"2024-07-29T20:36:57Z","created_at_i":1722285417,"num_comments":0,"objectID":"41103530","points":3,"story_id":41103530,"title":"\"Lambda: A Large Model Based Data Agent\"","updated_at":"2024-09-20T17:29:33Z","url":"https://github.com/Stephen-SMJ/LAMBDA"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"gravitino"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"What is Data Agent and how to build it in 15 Minutes"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://medium.com/datastrato/building-a-universal-data-agent-in-15-minutes-with-llamaindex-and-apache-gravitino-incubating-401ea24a3b39"}},"_tags":["story","author_gravitino","story_41266752"],"author":"gravitino","children":[41266753,41266767],"created_at":"2024-08-16T14:27:23Z","created_at_i":1723818443,"num_comments":1,"objectID":"41266752","points":2,"story_id":41266752,"title":"What is Data Agent and how to build it in 15 Minutes","updated_at":"2024-09-20T17:40:13Z","url":"https://medium.com/datastrato/building-a-universal-data-agent-in-15-minutes-with-llamaindex-and-apache-gravitino-incubating-401ea24a3b39"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"gagarwal123"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"We just posted results to the DataAgentBench leaderboard scoring 61.37% with GLM 5.2. Please check it out and do share your feedback"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Show HN: SQL MCP Server \u2013 61.37% on DataAgentBench with GLM-5.2"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/gagarwal304/databridge"}},"_tags":["story","author_gagarwal123","story_48656497","show_hn"],"author":"gagarwal123","created_at":"2026-06-24T07:37:24Z","created_at_i":1782286644,"num_comments":0,"objectID":"48656497","points":2,"story_id":48656497,"story_text":"We just posted results to the DataAgentBench leaderboard scoring 61.37% with GLM 5.2. Please check it out and do share your feedback","title":"Show HN: SQL MCP Server \u2013 61.37% on DataAgentBench with GLM-5.2","updated_at":"2026-06-24T09:53:17Z","url":"https://github.com/gagarwal304/databridge"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ashpreet-bedi"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Dash: The Data Agent Every Company Needs"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.ashpreetbedi.com/articles/dash-v2"}},"_tags":["story","author_ashpreet-bedi","story_47786107"],"author":"ashpreet-bedi","created_at":"2026-04-15T22:22:06Z","created_at_i":1776291726,"num_comments":0,"objectID":"47786107","points":2,"story_id":47786107,"title":"Dash: The Data Agent Every Company Needs","updated_at":"2026-05-07T20:45:46Z","url":"https://www.ashpreetbedi.com/articles/dash-v2"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dmpetrov"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"How to Build a Data Agent in 2026"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://twitter.com/jamiequint/status/2029705203457609785"}},"_tags":["story","author_dmpetrov","story_47279796"],"author":"dmpetrov","created_at":"2026-03-06T19:22:24Z","created_at_i":1772824944,"num_comments":0,"objectID":"47279796","points":2,"story_id":47279796,"title":"How to Build a Data Agent in 2026","updated_at":"2026-03-06T19:32:33Z","url":"https://twitter.com/jamiequint/status/2029705203457609785"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"theoradical"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Dash: A Self-Learning Data Agent That Remembers Its Mistakes"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://starlog.is/articles/ai-machine-learning/agno-agi-dash/"}},"_tags":["story","author_theoradical","story_47157408"],"author":"theoradical","created_at":"2026-02-25T20:30:17Z","created_at_i":1772051417,"num_comments":0,"objectID":"47157408","points":2,"story_id":47157408,"title":"Dash: A Self-Learning Data Agent That Remembers Its Mistakes","updated_at":"2026-03-05T23:37:22Z","url":"https://starlog.is/articles/ai-machine-learning/agno-agi-dash/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tamnd"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"OpenAI's In-House Data Agent"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://openai.com/index/inside-our-in-house-data-agent/?"}},"_tags":["story","author_tamnd","story_46868582"],"author":"tamnd","created_at":"2026-02-03T09:18:27Z","created_at_i":1770110307,"num_comments":0,"objectID":"46868582","points":2,"story_id":46868582,"title":"OpenAI's In-House Data Agent","updated_at":"2026-03-05T23:30:03Z","url":"https://openai.com/index/inside-our-in-house-data-agent/?"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ashpreetagno"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Dash: A self-learning data agent inspired by OpenAI's in-house data agent"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://github.com/agno-agi/data-agent"}},"_tags":["story","author_ashpreetagno","story_46831168"],"author":"ashpreetagno","created_at":"2026-01-30T23:03:17Z","created_at_i":1769814197,"num_comments":0,"objectID":"46831168","points":2,"story_id":46831168,"title":"Dash: A self-learning data agent inspired by OpenAI's in-house data agent","updated_at":"2026-03-05T23:27:33Z","url":"https://github.com/agno-agi/data-agent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"willahmad"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"OpenAI contract data agent \u2013 DocuGPT"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://openai.com/index/openai-contract-data-agent/"}},"_tags":["story","author_willahmad","story_45435247"],"author":"willahmad","created_at":"2025-10-01T07:26:14Z","created_at_i":1759303574,"num_comments":0,"objectID":"45435247","points":2,"story_id":45435247,"title":"OpenAI contract data agent \u2013 DocuGPT","updated_at":"2026-03-05T22:47:44Z","url":"https://openai.com/index/openai-contract-data-agent/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"aichelles"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Datafly \u2013 data agent that automatically understands any database you connect"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/dkeviv/datafly"}},"_tags":["story","author_aichelles","story_47330546"],"author":"aichelles","children":[47330547],"created_at":"2026-03-11T00:34:50Z","created_at_i":1773189290,"num_comments":3,"objectID":"47330546","points":1,"story_id":47330546,"title":"Datafly \u2013 data agent that automatically understands any database you connect","updated_at":"2026-03-11T03:51:45Z","url":"https://github.com/dkeviv/datafly"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"river_wu"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Data Agent Ready Database: Designing the Next-Gen Enterprise Data Warehouse"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.databend.com/blog/category-product/databend-agent-ready-database"}},"_tags":["story","author_river_wu","story_46881202"],"author":"river_wu","children":[46881203,46881223],"created_at":"2026-02-04T03:44:25Z","created_at_i":1770176665,"num_comments":1,"objectID":"46881202","points":1,"story_id":46881202,"title":"Data Agent Ready Database: Designing the Next-Gen Enterprise Data Warehouse","updated_at":"2026-03-05T23:30:59Z","url":"https://www.databend.com/blog/category-product/databend-agent-ready-database"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ashpreetagno"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Dash: Open-source implementation of OpenAI's in-house data agent"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.ashpreetbedi.com/articles/dash"}},"_tags":["story","author_ashpreetagno","story_46857031"],"author":"ashpreetagno","children":[46857032,46873768],"created_at":"2026-02-02T15:22:44Z","created_at_i":1770045764,"num_comments":1,"objectID":"46857031","points":1,"story_id":46857031,"title":"Dash: Open-source implementation of OpenAI's in-house data agent","updated_at":"2026-03-05T23:29:19Z","url":"https://www.ashpreetbedi.com/articles/dash"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mousematrix"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Faithful, Reproducible, Wrong: Lessons from 900 Data-Agent Runs"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://xorq.dev/blog/faithful-reproducible-wrong/"}},"_tags":["story","author_mousematrix","story_49437350"],"author":"mousematrix","created_at":"2026-08-25T17:13:12Z","created_at_i":1787677992,"num_comments":0,"objectID":"49437350","points":1,"story_id":49437350,"title":"Faithful, Reproducible, Wrong: Lessons from 900 Data-Agent Runs","updated_at":"2026-08-25T17:15:20Z","url":"https://xorq.dev/blog/faithful-reproducible-wrong/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ogham"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"How we built a chat-with-data agent that doesn't AI generate SQL"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://inconvo.com/blog/how-we-built-a-chat-with-data-agent-that-doesnt-ai-generate-sql/"}},"_tags":["story","author_ogham","story_47123226"],"author":"ogham","created_at":"2026-02-23T15:06:12Z","created_at_i":1771859172,"num_comments":0,"objectID":"47123226","points":1,"story_id":47123226,"title":"How we built a chat-with-data agent that doesn't AI generate SQL","updated_at":"2026-03-05T23:35:35Z","url":"https://inconvo.com/blog/how-we-built-a-chat-with-data-agent-that-doesnt-ai-generate-sql/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ReDeiPirati"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Adala: Autonomous Data Agent Framework"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/HumanSignal/Adala"}},"_tags":["story","author_ReDeiPirati","story_38208738"],"author":"ReDeiPirati","created_at":"2023-11-09T18:24:01Z","created_at_i":1699554241,"num_comments":0,"objectID":"38208738","points":1,"story_id":38208738,"title":"Adala: Autonomous Data Agent Framework","updated_at":"2024-09-20T15:33:22Z","url":"https://github.com/HumanSignal/Adala"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"twosdai"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Hey HN! Daniel here, I\u2019m a software engineer and hobbyist hacker. I\u2019m joined by my cofounder Matt. We\u2019re building Paigo (https://paigo.tech). We make it easy for SaaS businesses to bill customers based on usage.
To get your hands dirty a bit we have a stateless and signupless demo you can try out: https://hn.paigo.tech/ and a video of me walking through the system in a bit more detail: https://youtu.be/T6J1Yh8GhdU.
The idea of our platform is fairly straightforward: You give us read-only access to your SaaS backend and based on tenant metadata for your infrastructure, we measure, persist, and aggregate SaaS tenant usage data to give a clear picture of per-client usage. We can measure metrics like API requests, Compute time, Data Storage, Transaction Volumes and many more. Some common scenarios would be: an ML platform could use Paigo to track processed input files for customers, a Data platform could use Paigo to determine the data size customers have consumed, and an API company can use Paigo to track customers\u2019 API requests. Additionally, we also help you understand your cost to serve your clients\u2019 usage, and this data allows us to provide your SaaS with usage based billing.
What\u2019s the problem we are solving? Many SaaS products need to measure their customer's usage in some form, and many want to incorporate it into their billing plans. It\u2019s fairly annoying to either build the entire system in house or to build a measurement system in house and then connect to a billing provider. It takes months to get a usage based billing system up and running and usually requires several engineers (if not more) to maintain and operate. Also, when Sales wants to offer specific discounts or deals to major enterprises, it\u2019s typically handled outside of the in-house system in Excel spreadsheets with some good guesses. This is how a lot of money gets lost for major deals.
With Paigo we handle 100% of the measurement and collection of SaaS customers\u2019 usage for the business. SaaS business can see their customers\u2019 usage within 10 minutes, because all they need to do is give us read access to their cloud account. Since we pull the lower level infra-data we can additionally give information like per tenant cost, and profit margin.
Matt and I came to this project after we built similar internal billing systems at previous jobs and we realized how error-prone these systems can be\u2014one incident might have even undercharged a client by a few million dollars! We also realized there was no solution which integrated directly to a backend system and handled the measurement and gathering of usage data as well as providing the end billing integration to platforms like stripe, AWS marketplace, or through ACH.
To get into the technical details Paigo has a few measurement systems to measure different forms of usage data: infrastructure-based, where we connect directly to cloud APIs then to slice-and-dice per tenant usage data; agent-based, where our agent is deployed into a runtime to gather usage like pod cpu time, memory, and file read write, along with any exported metrics that are prometheus compatible; and datastore-based, where we connect directly to datastores like S3, Kinesis, or log file. We require that the data in the datastore based approach adhere to a standard data format so we can process it. However this allows us to Pull, any custom metrics and dimensions directly from your Datastore. All of this data is then processed and sent to our backend usage journal, where we store it in an append-only ledger pattern.
For clients to search, and aggregate their data into an end bill or to slice and dice their client\u2019s cost and usage we have an API clients can use. We\u2019re an API first company, which is why our demo can work with Retool\u2014the demo is just a very thin skin over our API. The API is a NestJS based application, currently running in AWS Lambda with API-Gateway.
We bill based on invoiced revenue (surprise surprise its usage based) and we have a platform fee, roughly it breaks down to 1% of invoiced revenue on Paigo. Note that pricing is not currently transparent on our website. Our typical customers are mid-sized enterprises where an initial sales call is typically expected. However, we will be updating our main webpage soon to have some self-service options.
For a bit of deeper dive on the measurement engine we have some docs here: https://docs.paigo.tech/
Thanks for taking time to read! Let us know what you hate and maybe what you love :P. We\u2019d also love to hear your thoughts and experiences with measuring customer usage and usage-based billing!"},"title":{"matchLevel":"none","matchedWords":[],"value":"Launch HN: Paigo (YC S22) \u2013 Measure and bill SaaS customers based on usage"}},"_tags":["story","author_twosdai","story_33333584","launch_hn"],"author":"twosdai","children":[33334083,33334221,33334347,33334427,33334576,33335553,33335731,33337223,33338548,33339240,33346396,33351713],"created_at":"2022-10-25T17:27:16Z","created_at_i":1666718836,"num_comments":35,"objectID":"33333584","points":89,"story_id":33333584,"story_text":"Hey HN! Daniel here, I\u2019m a software engineer and hobbyist hacker. I\u2019m joined by my cofounder Matt. We\u2019re building Paigo (https://paigo.tech). We make it easy for SaaS businesses to bill customers based on usage.
To get your hands dirty a bit we have a stateless and signupless demo you can try out: https://hn.paigo.tech/ and a video of me walking through the system in a bit more detail: https://youtu.be/T6J1Yh8GhdU.
The idea of our platform is fairly straightforward: You give us read-only access to your SaaS backend and based on tenant metadata for your infrastructure, we measure, persist, and aggregate SaaS tenant usage data to give a clear picture of per-client usage. We can measure metrics like API requests, Compute time, Data Storage, Transaction Volumes and many more. Some common scenarios would be: an ML platform could use Paigo to track processed input files for customers, a Data platform could use Paigo to determine the data size customers have consumed, and an API company can use Paigo to track customers\u2019 API requests. Additionally, we also help you understand your cost to serve your clients\u2019 usage, and this data allows us to provide your SaaS with usage based billing.
What\u2019s the problem we are solving? Many SaaS products need to measure their customer's usage in some form, and many want to incorporate it into their billing plans. It\u2019s fairly annoying to either build the entire system in house or to build a measurement system in house and then connect to a billing provider. It takes months to get a usage based billing system up and running and usually requires several engineers (if not more) to maintain and operate. Also, when Sales wants to offer specific discounts or deals to major enterprises, it\u2019s typically handled outside of the in-house system in Excel spreadsheets with some good guesses. This is how a lot of money gets lost for major deals.
With Paigo we handle 100% of the measurement and collection of SaaS customers\u2019 usage for the business. SaaS business can see their customers\u2019 usage within 10 minutes, because all they need to do is give us read access to their cloud account. Since we pull the lower level infra-data we can additionally give information like per tenant cost, and profit margin.
Matt and I came to this project after we built similar internal billing systems at previous jobs and we realized how error-prone these systems can be\u2014one incident might have even undercharged a client by a few million dollars! We also realized there was no solution which integrated directly to a backend system and handled the measurement and gathering of usage data as well as providing the end billing integration to platforms like stripe, AWS marketplace, or through ACH.
To get into the technical details Paigo has a few measurement systems to measure different forms of usage data: infrastructure-based, where we connect directly to cloud APIs then to slice-and-dice per tenant usage data; agent-based, where our agent is deployed into a runtime to gather usage like pod cpu time, memory, and file read write, along with any exported metrics that are prometheus compatible; and datastore-based, where we connect directly to datastores like S3, Kinesis, or log file. We require that the data in the datastore based approach adhere to a standard data format so we can process it. However this allows us to Pull, any custom metrics and dimensions directly from your Datastore. All of this data is then processed and sent to our backend usage journal, where we store it in an append-only ledger pattern.
For clients to search, and aggregate their data into an end bill or to slice and dice their client\u2019s cost and usage we have an API clients can use. We\u2019re an API first company, which is why our demo can work with Retool\u2014the demo is just a very thin skin over our API. The API is a NestJS based application, currently running in AWS Lambda with API-Gateway.
We bill based on invoiced revenue (surprise surprise its usage based) and we have a platform fee, roughly it breaks down to 1% of invoiced revenue on Paigo. Note that pricing is not currently transparent on our website. Our typical customers are mid-sized enterprises where an initial sales call is typically expected. However, we will be updating our main webpage soon to have some self-service options.
For a bit of deeper dive on the measurement engine we have some docs here: https://docs.paigo.tech/
Thanks for taking time to read! Let us know what you hate and maybe what you love :P. We\u2019d also love to hear your thoughts and experiences with measuring customer usage and usage-based billing!","title":"Launch HN: Paigo (YC S22) \u2013 Measure and bill SaaS customers based on usage","updated_at":"2024-09-20T12:22:48Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"micrum"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"You might know Cube as an open-source semantic layer (https://github.com/cube-js/cube). Started in 2018, now 19K+ stars, 1000+ releases.
We kept hitting the same wall: everyone wants AI analytics, but AI without business context hallucinates. The fix is a semantic layer \u2014 a model that defines what "revenue" or "churn" actually means. But building one by hand takes weeks.
So we built an AI agent that writes the semantic layer itself, then uses it to answer questions and build dashboards with no hallucinations.
Connect your data \u2192 agent builds the model in seconds \u2192 ask questions in natural language \u2192 get accurate results grounded in your business definitions.
Three months in GA. Running in 200+ companies. 500K+ lines of semantic layer code generated. Teams like Brex and Drata using it in production.\nOSS core isn't going anywhere.
Agentic analytics is a cloud layer on top. Free tier available. https://cube.dev/
The founders are in the thread to answer your question."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Cube \u2013 The Agentic Analytics Platform [video]"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.youtube.com/watch?v=f9RMT6WMAlc"}},"_tags":["story","author_micrum","story_46966941","show_hn"],"author":"micrum","children":[46967887],"created_at":"2026-02-10T21:07:28Z","created_at_i":1770757648,"num_comments":1,"objectID":"46966941","points":6,"story_id":46966941,"story_text":"You might know Cube as an open-source semantic layer (https://github.com/cube-js/cube). Started in 2018, now 19K+ stars, 1000+ releases.
We kept hitting the same wall: everyone wants AI analytics, but AI without business context hallucinates. The fix is a semantic layer \u2014 a model that defines what "revenue" or "churn" actually means. But building one by hand takes weeks.
So we built an AI agent that writes the semantic layer itself, then uses it to answer questions and build dashboards with no hallucinations.
Connect your data \u2192 agent builds the model in seconds \u2192 ask questions in natural language \u2192 get accurate results grounded in your business definitions.
Three months in GA. Running in 200+ companies. 500K+ lines of semantic layer code generated. Teams like Brex and Drata using it in production.\nOSS core isn't going anywhere.
Agentic analytics is a cloud layer on top. Free tier available. https://cube.dev/
The founders are in the thread to answer your question.","title":"Show HN: Cube \u2013 The Agentic Analytics Platform [video]","updated_at":"2026-03-05T23:34:17Z","url":"https://www.youtube.com/watch?v=f9RMT6WMAlc"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ygudeta"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"I built *valuepulse.ai* out of a frustration I kept running into: everything is scattered across different tools. If I want to search internal knowledge, I use something like Glean. If I want to analyze data, I switch to a BI tool or a data agent. And if I need external context, I\u2019m back to Google or Perplexity.
Even with all the progress in generative AI, it still feels pretty disconnected.
So I wanted to see if it\u2019s possible to bring search (both web and enterprise) and analytics into one place.
The core idea behind Valuepulse is a *context layer* that models a company\u2019s business domain (customers, revenue, operations, etc.) and sits across different data sources (structured and unstructured). On top of that, I built:
* natural language \u2192 dashboard generation (similar idea to tools like Replit/Lovable, but for BI), using agents like text-to-SQL and visualization selection\n* a data agent that can query databases and warehouses in plain English\n* unified search across internal docs and the web using vector search
It\u2019s still early (beta), but I\u2019d really appreciate feedback\u2014especially on whether this feels useful for real business users (non-technical) or where it might break."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Valuepulse \u2013 search docs, query data, and build dashboards in one place"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://valuepulse.ai"}},"_tags":["story","author_ygudeta","story_47802054","show_hn"],"author":"ygudeta","children":[47802997,47803138],"created_at":"2026-04-17T03:04:01Z","created_at_i":1776395041,"num_comments":1,"objectID":"47802054","points":2,"story_id":47802054,"story_text":"I built *valuepulse.ai* out of a frustration I kept running into: everything is scattered across different tools. If I want to search internal knowledge, I use something like Glean. If I want to analyze data, I switch to a BI tool or a data agent. And if I need external context, I\u2019m back to Google or Perplexity.
Even with all the progress in generative AI, it still feels pretty disconnected.
So I wanted to see if it\u2019s possible to bring search (both web and enterprise) and analytics into one place.
The core idea behind Valuepulse is a *context layer* that models a company\u2019s business domain (customers, revenue, operations, etc.) and sits across different data sources (structured and unstructured). On top of that, I built:
* natural language \u2192 dashboard generation (similar idea to tools like Replit/Lovable, but for BI), using agents like text-to-SQL and visualization selection\n* a data agent that can query databases and warehouses in plain English\n* unified search across internal docs and the web using vector search
It\u2019s still early (beta), but I\u2019d really appreciate feedback\u2014especially on whether this feels useful for real business users (non-technical) or where it might break.","title":"Show HN: Valuepulse \u2013 search docs, query data, and build dashboards in one place","updated_at":"2026-04-17T06:48:02Z","url":"https://valuepulse.ai"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"CShorten"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Hey everyone! I am SUPER EXCITED to publish a new episode of the Weaviate Podcast with Shreya Shankar on Data Agents!
Shreya is a Ph.D. student at UC Berkeley's EPIC Data Lab advised by Aditya Parameswaran. Her research focuses on advancing data systems and human-computer interaction!
This podcast dives into her latest work on the Data Agent Benchmark! This is the first benchmark testing how well agents can perform multi-step queries across multiple database systems!
We also covered DocETL and Semantic Operators, as well as how database principles can shape the future of AI agents, and why context management may be the new data management!
A lot of big takeaways from this one, I hope you find it useful!
YouTube: https://www.youtube.com/watch?v=C-fNVPYZrVg
Spotify: https://spotifycreators-web.app.link/e/juDmrVcp71b"},"title":{"matchLevel":"none","matchedWords":[],"value":"Data Agents with Shreya Shankar \u2013 Weaviate Podcast #135"}},"_tags":["story","author_CShorten","story_47661225","ask_hn"],"author":"CShorten","created_at":"2026-04-06T14:15:31Z","created_at_i":1775484931,"num_comments":0,"objectID":"47661225","points":1,"story_id":47661225,"story_text":"Hey everyone! I am SUPER EXCITED to publish a new episode of the Weaviate Podcast with Shreya Shankar on Data Agents!
Shreya is a Ph.D. student at UC Berkeley's EPIC Data Lab advised by Aditya Parameswaran. Her research focuses on advancing data systems and human-computer interaction!
This podcast dives into her latest work on the Data Agent Benchmark! This is the first benchmark testing how well agents can perform multi-step queries across multiple database systems!
We also covered DocETL and Semantic Operators, as well as how database principles can shape the future of AI agents, and why context management may be the new data management!
A lot of big takeaways from this one, I hope you find it useful!
YouTube: https://www.youtube.com/watch?v=C-fNVPYZrVg
Spotify: https://spotifycreators-web.app.link/e/juDmrVcp71b","title":"Data Agents with Shreya Shankar \u2013 Weaviate Podcast #135","updated_at":"2026-04-06T14:18:08Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tai-kimura"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"I built an ecosystem for AI-driven development where breaking architectural rules is structurally impossible.
*The problem:* AI coding assistants produce inconsistent code. Every session yields different implementations, and AI "forgets" rules mid-conversation. Prompt engineering helps, but quality still depends on how well you explain things each time.
*The insight:* Don't ask AI to follow rules\u2014make it impossible to break them.
*The approach:*
1. *Specialized agents with strict boundaries* - Instead of one AI doing everything, split responsibilities. Layout agent creates JSON UI structure (never touches data types). Data agent defines bindings (never writes business logic). ViewModel agent implements logic (never edits JSON).
2. *JSON as single source of truth* - One JSON definition generates iOS native (SwiftUI/UIKit), Android native (Compose/XML), Web (React/Tailwind), tests, and docs. All in sync. Always.
3. *Cross-platform test runner* - Same test JSON runs on XCUITest, UIAutomator, and Playwright.
*Result:* Spec, implementation, and docs stay in sync because they're generated from the same source. AI agents are productive because they have clear, narrow scopes.
Still in development. Repos:
- Core: SwiftJsonUI, KotlinJsonUI, ReactJsonUI\n- Test runner: jsonui-test-runner (CLI + platform drivers)\n- Agents: JsonUI-Agents-for-claude
GitHub: https://github.com/Tai-Kimura
Would love feedback on the agent design approach."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: JsonUI \u2013 Constrain AI agents through code structure, not prompts"}},"_tags":["story","author_tai-kimura","story_46754009","show_hn"],"author":"tai-kimura","created_at":"2026-01-25T13:43:14Z","created_at_i":1769348594,"num_comments":0,"objectID":"46754009","points":1,"story_id":46754009,"story_text":"I built an ecosystem for AI-driven development where breaking architectural rules is structurally impossible.
*The problem:* AI coding assistants produce inconsistent code. Every session yields different implementations, and AI "forgets" rules mid-conversation. Prompt engineering helps, but quality still depends on how well you explain things each time.
*The insight:* Don't ask AI to follow rules\u2014make it impossible to break them.
*The approach:*
1. *Specialized agents with strict boundaries* - Instead of one AI doing everything, split responsibilities. Layout agent creates JSON UI structure (never touches data types). Data agent defines bindings (never writes business logic). ViewModel agent implements logic (never edits JSON).
2. *JSON as single source of truth* - One JSON definition generates iOS native (SwiftUI/UIKit), Android native (Compose/XML), Web (React/Tailwind), tests, and docs. All in sync. Always.
3. *Cross-platform test runner* - Same test JSON runs on XCUITest, UIAutomator, and Playwright.
*Result:* Spec, implementation, and docs stay in sync because they're generated from the same source. AI agents are productive because they have clear, narrow scopes.
Still in development. Repos:
- Core: SwiftJsonUI, KotlinJsonUI, ReactJsonUI\n- Test runner: jsonui-test-runner (CLI + platform drivers)\n- Agents: JsonUI-Agents-for-claude
GitHub: https://github.com/Tai-Kimura
Would love feedback on the agent design approach.","title":"Show HN: JsonUI \u2013 Constrain AI agents through code structure, not prompts","updated_at":"2026-03-05T23:25:19Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"arsh9745774"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Rules that stop AI agents from wrecking your warehouse, with an eval"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://github.com/idk-arsh/data-agent-rules"}},"_tags":["story","author_arsh9745774","story_49902237","show_hn"],"author":"arsh9745774","children":[49902238],"created_at":"2026-09-29T23:25:20Z","created_at_i":1790724320,"num_comments":0,"objectID":"49902237","points":1,"story_id":49902237,"title":"Show HN: Rules that stop AI agents from wrecking your warehouse, with an eval","updated_at":"2026-09-30T00:31:39Z","url":"https://github.com/idk-arsh/data-agent-rules"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ogham"},"title":{"matchLevel":"none","matchedWords":[],"value":"We built chat-with-data without text-to-SQL"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://inconvo.com/blog/how-we-built-a-chat-with-data-agent-that-doesnt-ai-generate-sql/"}},"_tags":["story","author_ogham","story_47139963"],"author":"ogham","created_at":"2026-02-24T17:34:57Z","created_at_i":1771954497,"num_comments":0,"objectID":"47139963","points":1,"story_id":47139963,"title":"We built chat-with-data without text-to-SQL","updated_at":"2026-03-05T23:36:28Z","url":"https://inconvo.com/blog/how-we-built-a-chat-with-data-agent-that-doesnt-ai-generate-sql/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"vladkol"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Business Intelligence Agent"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://github.com/vladkol/crm-data-agent"}},"_tags":["story","author_vladkol","story_44075389","show_hn"],"author":"vladkol","created_at":"2025-05-23T18:47:22Z","created_at_i":1748026042,"num_comments":0,"objectID":"44075389","points":1,"story_id":44075389,"title":"Show HN: Business Intelligence Agent","updated_at":"2025-05-23T18:48:48Z","url":"https://github.com/vladkol/crm-data-agent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"anophelon"},"story_text":{"matchLevel":"none","matchedWords":[],"value":"Automate business hypothesis creation and testing"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: AI Powered Data Analyst"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://github.com/antonguzun/data-agent"}},"_tags":["story","author_anophelon","story_41975852","show_hn"],"author":"anophelon","created_at":"2024-10-28T20:27:46Z","created_at_i":1730147266,"num_comments":0,"objectID":"41975852","points":1,"story_id":41975852,"story_text":"Automate business hypothesis creation and testing","title":"Show HN: AI Powered Data Analyst","updated_at":"2024-10-28T20:31:58Z","url":"https://github.com/antonguzun/data-agent"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"finniananderson"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Fire declared in OVH SBG2 datacentre building"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://travaux.ovh.net/?do=details&id=49471&"}},"_tags":["story","author_finniananderson","story_26407323"],"author":"finniananderson","children":[26407494,26407609,26407613,26407634,26407661,26407682,26407698,26407717,26407764,26407794,26407815,26407847,26407891,26407908,26408128,26408183,26408206,26408237,26408256,26408277,26408293,26408390,26408437,26408439,26408498,26408508,26408523,26408529,26408578,26408600,26408612,26408653,26408678,26408691,26408723,26408774,26408817,26408826,26408839,26408841,26408890,26408925,26408938,26409094,26409112,26409227,26409341,26409368,26409388,26409480,26409560,26409603,26409621,26409640,26409662,26409692,26409737,26409797,26410053,26410102,26410159,26410236,26410347,26410395,26410509,26410567,26410763,26410764,26410910,26411071,26411131,26411213,26411378,26411551,26411576,26411666,26411840,26411988,26412118,26412141,26412363,26412453,26412772,26413597,26413715,26413897,26413901,26414771,26415930,26421696],"created_at":"2021-03-10T03:11:59Z","created_at_i":1615345919,"num_comments":589,"objectID":"26407323","points":1141,"story_id":26407323,"title":"Fire declared in OVH SBG2 datacentre building","updated_at":"2026-05-22T19:44:56Z","url":"http://travaux.ovh.net/?do=details&id=49471&"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ibobev"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Garage \u2013 An S3 object store so reliable you can run it outside datacenters"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://garagehq.deuxfleurs.fr/"}},"_tags":["story","author_ibobev","story_46326984"],"author":"ibobev","children":[46327432,46327434,46327636,46327771,46327927,46327940,46328056,46328182,46328218,46328246,46328515,46329097,46329372,46329378,46329405,46329640,46330003,46330070,46330201,46330319,46330440,46332539,46332957,46333377,46333897,46335154,46335446,46335784,46372612],"created_at":"2025-12-19T15:40:03Z","created_at_i":1766158803,"num_comments":182,"objectID":"46326984","points":722,"story_id":46326984,"title":"Garage \u2013 An S3 object store so reliable you can run it outside datacenters","updated_at":"2026-08-14T13:27:07Z","url":"https://garagehq.deuxfleurs.fr/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"martinald"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"xAI is looking more like a datacentre REIT than a frontier lab"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://martinalderson.com/posts/xais-new-rental-business/"}},"_tags":["story","author_martinald","story_48446428"],"author":"martinald","children":[48448431,48448498,48448520,48448859,48448896,48449126,48449154,48449216,48449378,48449539,48449557,48449579,48449688,48449826,48451359,48451921,48452764,48452866,48452954,48453054,48454251,48454514,48454549,48454627,48454918,48455306,48455673,48455715,48455762,48456293,48456403,48456487,48456613,48456687,48456962,48457518,48457767,48457778,48458354,48458414,48458740,48460352,48460893,48461031,48461090,48461189,48461958,48463595,48463693,48466703,48469399,48476614,48487176,48487282],"created_at":"2026-06-08T15:13:43Z","created_at_i":1780931623,"num_comments":552,"objectID":"48446428","points":692,"story_id":48446428,"title":"xAI is looking more like a datacentre REIT than a frontier lab","updated_at":"2026-07-24T09:32:37Z","url":"https://martinalderson.com/posts/xais-new-rental-business/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"baal80spam"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"AMD outsells Intel in the datacenter space"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://www.tomshardware.com/pc-components/cpus/for-the-first-time-ever-amd-outsells-intel-in-the-datacenter-space"}},"_tags":["story","author_baal80spam","story_42054449"],"author":"baal80spam","children":[42054997,42055008,42055016,42055193,42055201,42055684,42055984,42056147,42056387,42056669,42057337,42057338,42058043,42060555,42069650],"created_at":"2024-11-05T19:27:15Z","created_at_i":1730834835,"num_comments":161,"objectID":"42054449","points":499,"story_id":42054449,"title":"AMD outsells Intel in the datacenter space","updated_at":"2025-11-17T04:06:59Z","url":"https://www.tomshardware.com/pc-components/cpus/for-the-first-time-ever-amd-outsells-intel-in-the-datacenter-space"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mindracer"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Datacenters in space aren't going to work"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-good-idea/"}},"_tags":["story","author_mindracer","story_46087616"],"author":"mindracer","children":[46087941,46088005,46088020,46088037,46088051,46088060,46088072,46088098,46088239,46088352,46088467,46088479,46088616,46088619,46088629,46088900,46088911,46088977,46088999,46090446,46092258,46092857,46092981,46093006,46093103,46093230,46093294,46093306,46093339,46093425,46093475,46093507,46093705,46093910,46094171,46094236,46094792,46094793,46094848,46094936,46094944,46095111,46095249,46095320,46095534,46095547,46095790,46095838,46095947,46096212,46096243,46096644,46096694,46096828,46097032,46097080,46097127,46097441,46097729,46097838,46098635,46100197,46100230,46100293,46103393,46112795,46114339,46119291,46119294],"created_at":"2025-11-29T14:05:53Z","created_at_i":1764425153,"num_comments":437,"objectID":"46087616","points":498,"story_id":46087616,"title":"Datacenters in space aren't going to work","updated_at":"2026-06-07T14:33:27Z","url":"https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-good-idea/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"kristianp"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"We need a replacement for TCP in the datacenter [pdf]"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://web.stanford.edu/~ouster/cgi-bin/papers/replaceTcp.pdf"}},"_tags":["story","author_kristianp","story_33401480"],"author":"kristianp","children":[33401600,33401606,33401616,33401622,33401638,33401640,33401673,33401679,33401702,33401703,33401704,33401705,33401707,33401708,33401722,33401723,33401747,33401755,33401882,33401957,33401958,33402030,33402055,33402101,33402215,33402270,33402272,33402357,33402369,33402393,33402408,33402540,33402580,33402610,33402626,33402688,33402860,33402891,33402905,33403019,33403157,33403284,33403600,33403611,33403790,33403804,33403846,33403927,33403967,33403971,33404040,33404219,33404356,33404500,33404665,33404988,33405438,33405974,33406448,33406770,33409143,33409163,33409279,33412356,33417305,33417335],"created_at":"2022-10-31T02:41:43Z","created_at_i":1667184103,"num_comments":313,"objectID":"33401480","points":478,"story_id":33401480,"title":"We need a replacement for TCP in the datacenter [pdf]","updated_at":"2025-10-06T20:47:32Z","url":"https://web.stanford.edu/~ouster/cgi-bin/papers/replaceTcp.pdf"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mmastrac"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Shouting in the Datacenter (2008) [video]"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.youtube.com/watch?v=tDacjrSCeq4"}},"_tags":["story","author_mmastrac","story_32513240"],"author":"mmastrac","children":[32513684,32513749,32513877,32513933,32513967,32514028,32514130,32514195,32514242,32514365,32514424,32514428,32514477,32514540,32514570,32514758,32515300,32515809,32516304,32516898,32517522,32518374,32519729,32524494],"created_at":"2022-08-18T18:32:13Z","created_at_i":1660847533,"num_comments":78,"objectID":"32513240","points":437,"story_id":32513240,"title":"Shouting in the Datacenter (2008) [video]","updated_at":"2026-03-15T12:55:34Z","url":"https://www.youtube.com/watch?v=tDacjrSCeq4"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"f2n"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"NVIDIA GeForce driver deployment in datacenters is forbidden now"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://www.nvidia.com/content/DriverDownload-March2009/licence.php?lang=us&type=GeForce"}},"_tags":["story","author_f2n","story_16002068"],"author":"f2n","children":[16002098,16002257,16002260,16002274,16002286,16002323,16002327,16002351,16002379,16002380,16002402,16002403,16002437,16002450,16002457,16002480,16002506,16002513,16002519,16002524,16002565,16002631,16002833,16002911,16002966,16002982,16002988,16003046,16003076,16003196,16003239,16003277,16003322,16003451,16003541,16003675,16003705,16004101,16004860,16004999],"created_at":"2017-12-25T00:46:06Z","created_at_i":1514162766,"num_comments":200,"objectID":"16002068","points":358,"story_id":16002068,"title":"NVIDIA GeForce driver deployment in datacenters is forbidden now","updated_at":"2024-09-20T01:47:07Z","url":"http://www.nvidia.com/content/DriverDownload-March2009/licence.php?lang=us&type=GeForce"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"birdculture"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"I put a datacenter GPU in my gaming PC"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://blog.tymscar.com/posts/v100localllm/"}},"_tags":["story","author_birdculture","story_48345694"],"author":"birdculture","children":[48345889,48345897,48345923,48345925,48345930,48345938,48345961,48346026,48346034,48346038,48346046,48346048,48346069,48346183,48346217,48346302,48346381,48346386,48346392,48346457,48346467,48346485,48346746,48346923,48346968,48347222,48347653,48347952,48348203,48348228,48348424,48348443,48348847,48349567,48349946,48350178,48350653,48350663,48350777,48353283,48355783,48420558],"created_at":"2026-05-31T13:53:16Z","created_at_i":1780235596,"num_comments":192,"objectID":"48345694","points":332,"story_id":48345694,"title":"I put a datacenter GPU in my gaming PC","updated_at":"2026-08-17T21:48:18Z","url":"https://blog.tymscar.com/posts/v100localllm/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"sandwichsphinx"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Speed, scale and reliability: 25 years of Google datacenter networking evolution"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://cloud.google.com/blog/products/networking/speed-scale-reliability-25-years-of-data-center-networking"}},"_tags":["story","author_sandwichsphinx","story_42031169"],"author":"sandwichsphinx","children":[42031636,42031705,42031721,42031796,42032075,42032218,42032298,42032481,42032506,42032535,42032556,42032950,42033000,42033171,42034964,42036239],"created_at":"2024-11-03T04:29:14Z","created_at_i":1730608154,"num_comments":78,"objectID":"42031169","points":290,"story_id":42031169,"title":"Speed, scale and reliability: 25 years of Google datacenter networking evolution","updated_at":"2024-12-09T02:46:03Z","url":"https://cloud.google.com/blog/products/networking/speed-scale-reliability-25-years-of-data-center-networking"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"toddh"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Google reveals details about its datacenters"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"http://highscalability.com/blog/2015/8/10/how-google-invented-an-amazing-datacenter-network-only-they.html/"}},"_tags":["story","author_toddh","story_10037960"],"author":"toddh","children":[10039111,10039249,10039307,10039724,10039869,10040343,10040465,10041173,10041257,10042224],"created_at":"2015-08-10T22:43:14Z","created_at_i":1439246594,"num_comments":53,"objectID":"10037960","points":281,"story_id":10037960,"title":"Google reveals details about its datacenters","updated_at":"2023-09-06T23:21:04Z","url":"http://highscalability.com/blog/2015/8/10/how-google-invented-an-amazing-datacenter-network-only-they.html/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"zzzqqq"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Multi-Datacenter Cassandra on 32 Raspberry Pi\u2019s"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://www.datastax.com/dev/blog/32-node-raspberry-pi-cassandra-cluster"}},"_tags":["story","author_zzzqqq","story_8208316"],"author":"zzzqqq","children":[8208525,8208551,8208554,8208566,8208746,8208783,8208893,8208948,8208972,8209123,8209239,8209584,8211783],"created_at":"2014-08-21T16:57:07Z","created_at_i":1408640227,"num_comments":55,"objectID":"8208316","points":276,"story_id":8208316,"title":"Multi-Datacenter Cassandra on 32 Raspberry Pi\u2019s","updated_at":"2024-09-19T21:02:55Z","url":"http://www.datastax.com/dev/blog/32-node-raspberry-pi-cassandra-cluster"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dmit"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Haskell in the Datacentre"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://simonmar.github.io/posts/2016-12-08-Haskell-in-the-datacentre.html"}},"_tags":["story","author_dmit","story_13175040"],"author":"dmit","children":[13175176,13175236,13175341,13175375,13175548,13178036],"created_at":"2016-12-14T10:33:51Z","created_at_i":1481711631,"num_comments":43,"objectID":"13175040","points":276,"story_id":13175040,"title":"Haskell in the Datacentre","updated_at":"2024-09-20T00:08:01Z","url":"https://simonmar.github.io/posts/2016-12-08-Haskell-in-the-datacentre.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Bender"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Irish datacenters now guzzle 23% of the country's electricity"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://www.theregister.com/on-prem/2026/07/11/irish-datacenters-now-guzzle-23-of-the-countrys-electricity/5270013"}},"_tags":["story","author_Bender","story_48884322"],"author":"Bender","children":[48884475,48884571,48884584,48884641,48884739,48884857,48884880,48884891,48884914,48885022,48885026,48885070,48885096,48885264,48885269,48885316,48885332,48885377,48885449,48885609,48885765,48885893,48885982,48886045,48886171,48886476,48886700,48886848,48888015,48888601,48889964],"created_at":"2026-07-12T20:16:12Z","created_at_i":1783887372,"num_comments":309,"objectID":"48884322","points":247,"story_id":48884322,"title":"Irish datacenters now guzzle 23% of the country's electricity","updated_at":"2026-07-29T10:53:09Z","url":"https://www.theregister.com/on-prem/2026/07/11/irish-datacenters-now-guzzle-23-of-the-countrys-electricity/5270013"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Zaheer"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Photos from Bell Labs Datacenter in 1960s (2019)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://rarehistoricalphotos.com/inside-bell-labs-datacenter-1960s/"}},"_tags":["story","author_Zaheer","story_24105254"],"author":"Zaheer","children":[24105571,24105616,24105649,24105672,24105738,24105856,24105878,24105962,24105975,24105979,24106058,24106210,24106551,24107105,24107261,24107474,24107516,24109141,24109425,24110160,24110352,24110380,24113148,24116475,24117440],"created_at":"2020-08-10T03:38:00Z","created_at_i":1597030680,"num_comments":71,"objectID":"24105254","points":242,"story_id":24105254,"title":"Photos from Bell Labs Datacenter in 1960s (2019)","updated_at":"2024-09-20T06:42:32Z","url":"https://rarehistoricalphotos.com/inside-bell-labs-datacenter-1960s/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"davedx"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Are we repeating the telecoms crash with AI datacenters?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://martinalderson.com/posts/are-we-really-repeating-the-telecoms-crash-with-ai-datacenters/"}},"_tags":["story","author_davedx","story_46133141"],"author":"davedx","children":[46136217,46136301,46136429,46136461,46136494,46136532,46136536,46137061,46137094,46137186,46137556,46137612,46137731,46137777,46138103,46138322,46138645,46138685,46138933,46138986,46139039,46139359,46139415,46139473,46140054,46140613,46140701,46140898,46141380,46142723,46143407,46145009,46145149,46145399,46168830],"created_at":"2025-12-03T11:14:56Z","created_at_i":1764760496,"num_comments":194,"objectID":"46133141","points":241,"story_id":46133141,"title":"Are we repeating the telecoms crash with AI datacenters?","updated_at":"2026-09-27T09:58:31Z","url":"https://martinalderson.com/posts/are-we-really-repeating-the-telecoms-crash-with-ai-datacenters/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"el_duderino"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Microsoft to Deliver Microsoft Cloud from Datacenters in Africa"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"https://blogs.microsoft.com/blog/2017/05/18/microsoft-deliver-microsoft-cloud-datacenters-africa/"}},"_tags":["story","author_el_duderino","story_14367205"],"author":"el_duderino","children":[14367783,14367889,14367896,14368179,14368190,14368312,14368585,14368600,14368890,14368976,14369326,14369556,14370288,14370289,14371355,14371762,14373857,14373886],"created_at":"2017-05-18T13:56:11Z","created_at_i":1495115771,"num_comments":110,"objectID":"14367205","points":208,"story_id":14367205,"title":"Microsoft to Deliver Microsoft Cloud from Datacenters in Africa","updated_at":"2026-07-09T20:20:47Z","url":"https://blogs.microsoft.com/blog/2017/05/18/microsoft-deliver-microsoft-cloud-datacenters-africa/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"creativityhurts"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"Photo tour of Facebook\u2019s new datacenter"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["dataagent"],"value":"http://scobleizer.com/2011/04/16/photo-tour-of-facebooks-new-datacenter/"}},"_tags":["story","author_creativityhurts","story_2453734"],"author":"creativityhurts","children":[2453792,2453800,2453812,2453814,2453869,2453892,2454003,2454006,2454092,2454094,2454130,2454557,2454728,2454901,2454953,2455715,2456348],"created_at":"2011-04-16T10:59:49Z","created_at_i":1302951589,"num_comments":49,"objectID":"2453734","points":208,"story_id":2453734,"title":"Photo tour of Facebook\u2019s new datacenter","updated_at":"2023-09-06T19:40:48Z","url":"http://scobleizer.com/2011/04/16/photo-tour-of-facebooks-new-datacenter/"}],"hitsPerPage":50,"nbHits":3835,"nbPages":20,"page":0,"params":"query=DataAgent&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":12,"processingTimingsMS":{"_request":{"roundTrip":15},"afterFetch":{"format":{"highlighting":1,"total":1}},"fetch":{"query":6,"scanning":3,"total":10},"total":12},"query":"DataAgent","serverTimeMS":14}