Tinybird Code is a CLI agent with deep ClickHouse knowledge. Use it to develop, deploy, iterate, optimize, and scale your real-time data pipelines - from idea to production and beyond.
Hey hunters! We're excited to announce Tinybird Code, a CLI agent that acts as your trusted AI ClickHouse expert. It's perfect for building real-time analytics backends from scratch, writing ClickHouse SQL queries, exploring large amounts of real-time data, or deploying and optimizing pipelines in production.
@cameron_archer Congrats on the launch!! Can't wait to give this a try, I already have a project in mind for it š
I think it could be interesting if there was a way for Claude Code to call Tinybird Code as an agent. Eg. I'm using CC to build a larger app / project, and I want it to use Tinybird Code as an agent whenever it needs to do something related to analytics. Does that make any sense(?)
Real-time analytics APIs up in minutes? Thatās wildāno more wrestling with messy infra. This is gonna save devs sooo much pain, tbh. Mad respect to the team!
tried the interactive demo but found it a bit limiting when testing more complex joins and nested queries. Not sure if it's just the demo version or something in the tool itself. Either way, looking forward to improvement
Massive win for data teams ā Tinybird Code feels like having a senior ClickHouse engineer on call 24/7. Game-changing for building and scaling real-time analytics fast.
The ability to generate mock data and automated test suites seems particularly useful for building reliable data pipelines. Are there any features to help validate query performance before deploying to production? @rafamoreno
Replies
Tinybird
Hey hunters! We're excited to announce Tinybird Code, a CLI agent that acts as your trusted AI ClickHouse expert. It's perfect for building real-time analytics backends from scratch, writing ClickHouse SQL queries, exploring large amounts of real-time data, or deploying and optimizing pipelines in production.
I recorded a more comprehensive demo (14 minutes) here -> https://youtu.be/TBsccGglCq0
Basically, any of the heavy lifting you'd need to hire a ClickHouse engineer for, Tinybird Code can do for you, including:
⬩Defining table schemas, writing SQL, and building query lambdas (APIs)
⬩Analyzing performance logs to optimize table schemas and SQL queries
⬩Iterating queries for better performance / lower cost
⬩Changing table schemas and migrating production data
⬩Generating mock data and building test suites
⬩Deploying projects to prod
Here's some more resources:
⬩Interactive Demo
⬩Docs
Ghost
@cameron_archer Congrats on the launch!! Can't wait to give this a try, I already have a project in mind for it š
I think it could be interesting if there was a way for Claude Code to call Tinybird Code as an agent. Eg. I'm using CC to build a larger app / project, and I want it to use Tinybird Code as an agent whenever it needs to do something related to analytics. Does that make any sense(?)
Tinybird
@johnonolan you're right on the money.
Check this out: https://x.com/tinybirdco/status/1948035650390819175
More content coming on this, soon ;)
BestPage.ai
Real-time analytics APIs up in minutes? Thatās wildāno more wrestling with messy infra. This is gonna save devs sooo much pain, tbh. Mad respect to the team!
tried the interactive demo but found it a bit limiting when testing more complex joins and nested queries. Not sure if it's just the demo version or something in the tool itself. Either way, looking forward to improvement
Tinybird
@kashmeer_rathour the interactive demo is just a marketing page to show the features - it doesn't actually have any functionality.
Did you try actually running the agent?
Massive win for data teams ā Tinybird Code feels like having a senior ClickHouse engineer on call 24/7. Game-changing for building and scaling real-time analytics fast.
Smoopit
The ability to generate mock data and automated test suites seems particularly useful for building reliable data pipelines. Are there any features to help validate query performance before deploying to production? @rafamoreno