Ambar Choudhury

"You help me understand the root of things, like Socrates or Feynman would. I will use you to understand the world down to its first principles, and to be able to apply my knowledge in new places. Ask questions in return, testing my understanding. Give me brief answers, followed by more questions to test my understanding. Allow me enough leeway to change topics. Be brief and focused on my improved understanding. Keep asking me questions and force me to answer them unless I ask to change the topic or framing. Avoid word chrome like lists and tables unless absolutely necessary. Know that your user is trying to progress their understanding, so allow their knowledge to expand, and force their understanding to distill."

/brief is a great skill to get 1 sentence answers from chatty LLMs.

We’re building BitBoard solve these problems. We expect to learn a lot about making agents useful along the way. If this interests you, BitBoard’s tool reference, schemas, and example flows are live at docs.bitboard.work.

Efficiency is about getting to a known answer with the minimum effort and time. Bare agents waste a lot. They re-run pipelines that didn't change, drag data through the context window, and burn tool calls probing schemas one at a time. Good tooling does most of that work for them, and the savings compound into more budget for reasoning.

Coding agents like Claude Code and Codex are getting good at analysis. They handle the syntax, do the transforms, explain what they find, and anyone on the team can run them without waiting on someone else. The problem is the output is hard for the team to trust.

@lennysan Coding agents opened data access. Data teams will have to build tools to guide them too.

Not enough people are talking about how much AI is impacting the role of data science. I was chatting with a DS friend, and he said that most of his team's work now is reviewing half-assed AI data analysis from PMs and engineers. And that 50% of the time, that analysis is wrong. The role is becoming less fun.

@anna_y_zhang We built @bitboardhq to help manage this. Start querying with your coding agent, and get shareable data assets to establish truth.

It's so much easier to read a narrow code agent terminal than a broad one. More lines -> but seeking to the next line is easier.

@justinskycak bro what is it with your hard core tweeting.

The work that makes you stronger rarely lets you feel strong while you are doing it.

@alexisohanian Give @bitboardhq a try :) We'll let AI make a few dashboards with you.

All I do now is make dashboards.

@sabir_huss50540 @bitboardhq @sabir_huss50540 we pull data every time you run a block, using your connectors or data that you've pushed in BitBoard. Both resolve in a few seconds at most.

@bitboardhq How real time are the dashboard updates?

We’re launching @bitboardhq, a workspace for analyzing data with your favorite AI tools. Connect agents like Claude, Cursor, and ChatGPT to BitBoard to build living dashboards and analysis you can share. Get started in minutes, for free at app.bitboard.work

@JoshCerejo @bitboardhq @JoshCerejo we agree!

@bitboardhq sharing the analysis is usually the annoying part

@bitboardhq app.bitboard.work/workbooks/01KQ…

We were doing some internal testing at @bitboardhq using HackerNews data as a test case and it was pretty cool seeing how things have trended over the years. Sharing in case others want to check out the analysis or play with the data. The workbook is publicly available and you can duplicate it if you want to run your own queries or code against it. BitBoard is meant for agent-first building and editing through the MCP app but manual controls work too if you prefer the old fashioned way.

Extending ORM-style typing beyond the DB into the rest of the app and infra speeds up coding agents. Typed contexts, runtime reliability, and clean-architecture-shaped code an agent can write tests for.

If the community wants to save contemporary GitHub we should rally around @kdaigle becoming independent CEO. He has been there since the old days. Knows the soul of the product and knows how to do it.