Javi Santana
@javisantana.bsky.social
Tinybird founder
Data infra for developers
Tinybird.co
Data infra for developers
Tinybird.co
Why we maintain a ClickHouse fork at Tinybird (and how it's different)
www.tinybird.co/blog-posts/w...
www.tinybird.co/blog-posts/w...
Why we maintain a ClickHouse fork at Tinybird (and how it's different)
We've long contributed to the open source ClickHouse community, and we didn't make this decision lightly. Here's why we maintain our own private ClickHouse fork.
www.tinybird.co
September 11, 2025 at 2:04 PM
Why we maintain a ClickHouse fork at Tinybird (and how it's different)
www.tinybird.co/blog-posts/w...
www.tinybird.co/blog-posts/w...
I wrote about how to ingest 1b rows per second with ClickHouse.
The point of the post is not the number, pushing 1b rows is “easy” the actual challenge comes when you need to do that in a reliable way, that’s what I try to explain
The point of the post is not the number, pushing 1b rows is “easy” the actual challenge comes when you need to do that in a reliable way, that’s what I try to explain
August 18, 2025 at 5:00 PM
I wrote about how to ingest 1b rows per second with ClickHouse.
The point of the post is not the number, pushing 1b rows is “easy” the actual challenge comes when you need to do that in a reliable way, that’s what I try to explain
The point of the post is not the number, pushing 1b rows is “easy” the actual challenge comes when you need to do that in a reliable way, that’s what I try to explain
I'm fucking tired of clickbaity content, people writing bad content just for SEO or not going into details because "people have a short attention span". Maybe that's true but some people appreciate long-form, detailed and super technical content.
Today we posted 5 high quality engineer posts:
Today we posted 5 high quality engineer posts:
June 27, 2025 at 11:46 AM
I'm fucking tired of clickbaity content, people writing bad content just for SEO or not going into details because "people have a short attention span". Maybe that's true but some people appreciate long-form, detailed and super technical content.
Today we posted 5 high quality engineer posts:
Today we posted 5 high quality engineer posts:
The second part of "handling clickhouse clusters at petabyte scale" www.tinybird.co/blog-posts/w...
Lessons learned from 5 years operating huge ClickHouse® clusters: Part II
This is the second part of the series. Here's more of what I've learned from operating petabyte-scale ClickHouse clusters for the last 5+ years.
www.tinybird.co
April 17, 2025 at 7:37 AM
The second part of "handling clickhouse clusters at petabyte scale" www.tinybird.co/blog-posts/w...
We've been operating petabyte-scale cluster for 5 years now, especially dealing with low latency use cases with large amounts of data, some of them under 10ms (for reference, most big data systems latency is 1 second at best).
www.tinybird.co/blog-posts/w...
www.tinybird.co/blog-posts/w...
I've operated petabyte-scale ClickHouse® clusters for 5 years
I've been operating large ClickHouse clusters for years. Here's what I've learned about architecture, storage, upgrades, config, testing, costs, and ingestion.
www.tinybird.co
April 2, 2025 at 1:59 PM
We've been operating petabyte-scale cluster for 5 years now, especially dealing with low latency use cases with large amounts of data, some of them under 10ms (for reference, most big data systems latency is 1 second at best).
www.tinybird.co/blog-posts/w...
www.tinybird.co/blog-posts/w...
I'd appreciate if you could upvote us in ProductHunt. It's about our latest product iteration
www.producthunt.com/posts/tinybi...
www.producthunt.com/posts/tinybi...
Tinybird Forward » - The fastest way to ship software with big data requirements | Product Hunt
Tinybird Forward is a new evolution of the Tinybird user experience. Build software with big data requirements faster and with fewer dependencies. Test locally, bootstrap with a prompt, deploy with on...
www.producthunt.com
March 18, 2025 at 1:28 PM
I'd appreciate if you could upvote us in ProductHunt. It's about our latest product iteration
www.producthunt.com/posts/tinybi...
www.producthunt.com/posts/tinybi...
Reposted by Javi Santana
🌅 Tinybird Forward is here »
Forward is a major evolution of Tinybird, designed to make shipping software with big data requirements faster and more intuitive.
No complex infra project. No context switching. No esoteric architectures. Just code.
(🔊 sound on! 👇)
youtu.be/vaSjWu3XFdY
Forward is a major evolution of Tinybird, designed to make shipping software with big data requirements faster and more intuitive.
No complex infra project. No context switching. No esoteric architectures. Just code.
(🔊 sound on! 👇)
youtu.be/vaSjWu3XFdY
Introducing Tinybird Forward: Ship software with big data requirements.
YouTube video by Tinybird
youtu.be
March 14, 2025 at 12:59 PM
🌅 Tinybird Forward is here »
Forward is a major evolution of Tinybird, designed to make shipping software with big data requirements faster and more intuitive.
No complex infra project. No context switching. No esoteric architectures. Just code.
(🔊 sound on! 👇)
youtu.be/vaSjWu3XFdY
Forward is a major evolution of Tinybird, designed to make shipping software with big data requirements faster and more intuitive.
No complex infra project. No context switching. No esoteric architectures. Just code.
(🔊 sound on! 👇)
youtu.be/vaSjWu3XFdY
Production hurts.
If it doesn't, it means you didn't try hard enough.
If it doesn't, it means you didn't try hard enough.
March 14, 2025 at 11:00 AM
Production hurts.
If it doesn't, it means you didn't try hard enough.
If it doesn't, it means you didn't try hard enough.
A new LLM version release with a new super powerful option: --schema
An mini BI example, in this case I'm querying a parquet file with all the events coming from gitlab tickets webhook
An mini BI example, in this case I'm querying a parquet file with all the events coming from gitlab tickets webhook
March 1, 2025 at 7:16 AM
A new LLM version release with a new super powerful option: --schema
An mini BI example, in this case I'm querying a parquet file with all the events coming from gitlab tickets webhook
An mini BI example, in this case I'm querying a parquet file with all the events coming from gitlab tickets webhook
Data intensive SaaS products will love to leave your data in icerbeg/s3
Guess who is going to pay all the storage and operation costs? 🤣
Guess who is going to pay all the storage and operation costs? 🤣
February 21, 2025 at 7:25 AM
Data intensive SaaS products will love to leave your data in icerbeg/s3
Guess who is going to pay all the storage and operation costs? 🤣
Guess who is going to pay all the storage and operation costs? 🤣
We’ve got a deep technical post almost ready on real-time log analytics, based on our experience designing these systems with companies handling trillions of rows.
February 19, 2025 at 10:11 AM
We’ve got a deep technical post almost ready on real-time log analytics, based on our experience designing these systems with companies handling trillions of rows.
We are using LLMs in the next Tinybird iterations but we are trying to be subtle, not going with the conversational interface. So for example, when you want to create synthetic data to test your logic you can do
𝚝𝚋 𝚖𝚘𝚌𝚔 𝚝𝚊𝚋𝚕𝚎
𝚝𝚋 𝚖𝚘𝚌𝚔 𝚝𝚊𝚋𝚕𝚎
February 13, 2025 at 11:40 AM
We are using LLMs in the next Tinybird iterations but we are trying to be subtle, not going with the conversational interface. So for example, when you want to create synthetic data to test your logic you can do
𝚝𝚋 𝚖𝚘𝚌𝚔 𝚝𝚊𝚋𝚕𝚎
𝚝𝚋 𝚖𝚘𝚌𝚔 𝚝𝚊𝚋𝚕𝚎
I though the LLM and IA thing was going to open a new world for UI design but the reality is that we are all doing chat interfaces
Maybe that’s the right thing to do and my expectations were too high
Maybe that’s the right thing to do and my expectations were too high
February 13, 2025 at 9:43 AM
I though the LLM and IA thing was going to open a new world for UI design but the reality is that we are all doing chat interfaces
Maybe that’s the right thing to do and my expectations were too high
Maybe that’s the right thing to do and my expectations were too high
About Iceberg C++ implementations. There are already some of them working (Redpanda Data and StarRocks seem to have good implementations for writing).
February 11, 2025 at 11:29 AM
About Iceberg C++ implementations. There are already some of them working (Redpanda Data and StarRocks seem to have good implementations for writing).
Reposted by Javi Santana
I wanted to understand how agents work, so I created a super simple one to run SQL queries. I wrote about the process
failingwithdata.substack.com/p/building-a...
failingwithdata.substack.com/p/building-a...
Building a simple SQL Agent from Scratch
with just bash + llm
failingwithdata.substack.com
January 28, 2025 at 9:16 AM
I wanted to understand how agents work, so I created a super simple one to run SQL queries. I wrote about the process
failingwithdata.substack.com/p/building-a...
failingwithdata.substack.com/p/building-a...
I always like to go to basics and learn from them. Using libraries, frameworks and tooling is perfect but understanding why things happen helps make decisions when designing software.
January 28, 2025 at 9:16 AM
I always like to go to basics and learn from them. Using libraries, frameworks and tooling is perfect but understanding why things happen helps make decisions when designing software.
I made a mistake running a SQL agent, I was sending the wrong data and I noticed that after some iterations it gets passive-aggressive:
* This is the final time I will tell you: I need the SQL query. You are providing the error message, not the SQL code.
* This is the final time I will tell you: I need the SQL query. You are providing the error message, not the SQL code.
January 23, 2025 at 9:00 AM
I made a mistake running a SQL agent, I was sending the wrong data and I noticed that after some iterations it gets passive-aggressive:
* This is the final time I will tell you: I need the SQL query. You are providing the error message, not the SQL code.
* This is the final time I will tell you: I need the SQL query. You are providing the error message, not the SQL code.
Reposted by Javi Santana
Building software that works is very easy… as long as you ignore all the hard parts.
January 19, 2025 at 2:01 PM
Building software that works is very easy… as long as you ignore all the hard parts.
Why would someone create your data pipeline tests by hand?
When your pipelines have parameters and you need to check different behaviors creating tests for the different paths is a pain.
When your pipelines have parameters and you need to check different behaviors creating tests for the different paths is a pain.
January 17, 2025 at 4:47 PM
Why would someone create your data pipeline tests by hand?
When your pipelines have parameters and you need to check different behaviors creating tests for the different paths is a pain.
When your pipelines have parameters and you need to check different behaviors creating tests for the different paths is a pain.
Simple question for people leading technical teams (hard answer tho): how do you create a writing culture in your company/team?
January 17, 2025 at 9:11 AM
Simple question for people leading technical teams (hard answer tho): how do you create a writing culture in your company/team?
When you have a data project and you want your editor to play well with it, you can generate all the context cursor needs.
January 16, 2025 at 2:30 PM
When you have a data project and you want your editor to play well with it, you can generate all the context cursor needs.
How I’d design the core of a cloud native database OLAP database (what Warpstream made with Kafka):
- Storage: S3 tables
- Short lived queries and ingestion: AWS lambda
- Batch jobs: Spot EC2 instances
Plus a vectorized execution engine.
- Storage: S3 tables
- Short lived queries and ingestion: AWS lambda
- Batch jobs: Spot EC2 instances
Plus a vectorized execution engine.
January 14, 2025 at 3:09 PM
How I’d design the core of a cloud native database OLAP database (what Warpstream made with Kafka):
- Storage: S3 tables
- Short lived queries and ingestion: AWS lambda
- Batch jobs: Spot EC2 instances
Plus a vectorized execution engine.
- Storage: S3 tables
- Short lived queries and ingestion: AWS lambda
- Batch jobs: Spot EC2 instances
Plus a vectorized execution engine.
Reposted by Javi Santana
I have a general one and another for Python that teaches cursor write PEP 735 compatible scripts!
Shared them here in a recent post.
davidgasquez.com/cursor-agent...
Shared them here in a recent post.
davidgasquez.com/cursor-agent...
January 13, 2025 at 7:34 AM
I have a general one and another for Python that teaches cursor write PEP 735 compatible scripts!
Shared them here in a recent post.
davidgasquez.com/cursor-agent...
Shared them here in a recent post.
davidgasquez.com/cursor-agent...