←
Raúl Ochoa133 posts
Raúl Ochoa avatar

Raúl Ochoa

@rochoa

Speed matters. Co-founded @tinybird.

◷ Joined October 2006▧ 6 Photos and videos
Javier Álvarez Medina avatar
⟳ Raúl Ochoa retweeted
Javier Álvarez Medina@xavijam·Oct 2

After being away from LinkedIn (well, everywhere) for several years, I’m catching up. What makes me happiest is seeing people I met at Tinybird building their own startups or doing incredible things at top companies. It’s such a joy to come across these stories.

Raúl Ochoa avatar
Raúl Ochoa@rochoa·Mar 4

I’m exploring new opportunities. My interest is in data infrastructure, high-concurrency, and technical leadership. I like building things that shouldn't work at scale, but do. If you're building a technical org or solving hard data problems, my DMs are open. No rush, just looking for the right fit.

Raúl Ochoa avatar
Raúl Ochoa@rochoa·Mar 4

After almost 7 years, I am no longer with Tinybird. I spent that time building the company from the initial versions of the product to a platform processing trillions of rows and petabytes of data on ClickHouse. Scaling the tech was a great challenge, but building and working with the engineering team behind it was the best part.

Marc Lou avatar
⟳ Raúl Ochoa retweeted
Marc Lou@marclou·25 Apr 2025

This is one of the best metered pricing UX I've seen 🤩 @tinybird shows the number of CPUs you get on the scroll label. That's nice!

0:07
Steven Tey avatar
⟳ Raúl Ochoa retweeted
Steven Tey@steventey·14 Mar 2025

Tinybird is the infra backbone for @dubdotco Analytics, helping us go from 0 → 250M clicks in the last 2 years. We're now doing ~75M clicks/mo and with that, it's critical that any DB migrations we make don't break prod. Tinybird's code-first DX is a game-changer for this.

Tinybird@tinybird14 Mar 2025

🌅 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. Here's what's new: » Local-first developer experience » Instant feedback loops during local dev » Native integration with AI coding agents » End-to-end testing in CI/CD without cloud » Single-command deploy/rollback » Frictionless schema migrations in production » Self-managed Tinybird deployments, for free (🔊 sound on! 👇)

2:54
Raúl Ochoa avatar
Raúl Ochoa@rochoa·18 Jul 2024

Just a reminder: on Azure, data transfer across availability zones using private or public IPs is free.

Post attachment
Raúl Ochoa avatar
Raúl Ochoa@rochoa·5 Jun 2024

What do we want to do? Roll out new ClickHouse versions as we release any other software in our stack. The flow should be like this: create a pull request with the target ClickHouse version, run the tests in the CI, merge it, deploy it, repeat.

Raúl Ochoa avatar
Raúl Ochoa@rochoa·5 Jun 2024

Of course, this sounds simpler than it is. Although many of the processes are automated and we have tooling around them, there are many decisions that are still human-driven. For example, what is the next cluster to upgrade? Do we upgrade to this or that version?

Raúl Ochoa avatar
Raúl Ochoa@rochoa·5 Jun 2024

For users' queries, we also perform automatic testing, running queries from our logs against ClickHouse instances that mimic production schemas and data types. We run these tests to ensure queries from the older versions are compatible with the latest versions.