Two weeks ago, we published v1 of our LLM SQL Generation Benchmark. Your feedback was generous, thoughtful, and sometimes brutally honest. Here's what we got wrong, what we got right (but didn’t explain), and how we’re improving the benchmark for round two. tbrd.co/llm-sql-rd2

One of the most popular crypto wallets in the world, Phantom, builds features with Tinybird to increase revenue and swaps by displaying trending tokens to end-users in milliseconds. Top quote from their Sr. Data Engineer ↓

Get to know your data's data. Exploratory Data Analysis in Tinybird -> tbrd.co/eda

How could I have missed it @tinybird.co just solved the biggest pain point to self host @openstatus.dev with their local container 🥰 www.tinybird.co/docs/forward...

Having fun with @shadcn.com sidebar. Inspired by @tinybird.co.

Introducing Explorations, a chat UI for real-time analytics. youtu.be/pLI1xLxpUTw

dbt shook up the data world in 2016. We like dbt. But it's not suited for real-time workloads (because data warehouses aren't suited for real-time workloads). Tinybird is a lot like dbt. But it's also different... 👀 tbrd.co/dbt-real-time

We just wrapped a blog series on counting 100 billion rows efficiently. The takeaway? Tinybird is fast and cost-effective when it comes to real-time aggregation at scale. Duh. Links to all 3 below ↓

Quick update on this ↓ Spots are filling up. We're ~75% full with a week to go. If you want to meet other #NewYorkCity devs building real-time data applications and learn from someone who has actually built and scaled the thing in prod, register now. Register: lu.ma/9wazu8zx

Where my Copenhagen devs at? Next Wednesday we're running a Tinybird Hackathon at the @BlastTV offices in Copenhagen 🇩🇰. Learn how Blast built their stats leaderboard for the game Deadlock, and get hands-on experience building real-time analytics APIs. Register here: lu.ma/wafkvnza

There are 6 functions to get a unique count in Tinybird: 1. uniq 2. uniqExact 3. uniqCombined 4. uniqCombined64 5. uniqHLL2 6. uniqTheta Each one has tradeoffs. Read how to combine them to efficiently count billions of unique IDs. ↓

💡 Quick win to improve the UX of your real-time analytics dashboards: Add an "Ask AI" feature. ↓ More info on how to do it in the 🧵

Just announced: Real-time Data Meetup with Tinybird, Estuary, and PlayOn! Sports. 3 talks. A room full of devs and data people. Free food and imbibements. April 29th at 6 PM ET FirstMark Capital - NYC Register here: lu.ma/9wazu8zx

Things that should be simple, but aren't: Canceling a database query from a web app client. Here's how we do it ⤑ tbrd.co/cancel-query

Own your data. Build your dashboard. A few months ago, we released a @vercel.com edge pinger to monitor your endpoints and store the responses in @tinybird.co: a lightweight OpenStatus version for you to self-host. Today, the UI is here. → logs.run/light

Building a real-time inventory system that actually works is harder than it looks. A thread on why 👇

You should instrument your application's LLM calls so you can analyze cost, usage, and adoption over time. Here are some example implementations in Python and TypeScript: tbrd.co/inst-llm

Vibe code tools, not toys.

The Vibeathon is on!!! Check out the prompt -> tbrd.co/vibeathon Commit your code by 12:48 PM ET. Win $3k! Va va va!!!!

The Tinybird Forward Vibeathon begins in 1 hr. Build something in 48 minutes, win $3k. A few tips to prepare: 💡 Install the latest Tinybird CLI 💡 Start tb local 10 min. before 💡 Clone the repo (link below) 💡 Start prepping your prompts! Here's the repo -> tbrd.co/vibeathon Good luck!

How to count 100,000,000,000 unique IDs in <100ms 👇 tbrd.co/100b-pt1

Introducing Tinybird Self-Managed. Run Tinybird on your own infra (for free). youtu.be/kBsSoaElRCo

Ask us anything about Tinybird Forward, starting in 15 minutes. gcal invite -> tbrd.co/fwd-ama youtube -> youtube.com/live/NIdIarNwm7U

You had me at "automatic schema migrations." Data deployments are no longer a mess of manual steps, risky schema changes & crossed fingers. With tb deploy, you get true CI/CD for your data - validation, migrations, backfills, and zero-downtime promotion in a SINGLE command. youtu.be/P3ZAgSlFJuQ

Block off your calendar for the Tinybird Forward Vibeathon. Up for grabs: Over $3,000 in credits from Tinybird, Vercel, Resend, Dub, Plain, and more. Wednesday, March 26th from 12:00 - 12:48 PM ET. Add it to your google cal👇

Writing tests sucks. Use LLMs to make it suck less.

Use tb mock && tb test to generate production-like mock data and a suite of unit tests for your APIs.

`tb dev` is the instant feedback loop for building your analytics backend. It's like `next dev`, but for data. Change your queries or change your data, and TInybird hot reloads your build in milliseconds. youtu.be/DCzYj9VKdHE

Tinybird just launched Forward. Curious how it works? Ready to migrate? Want a feature rundown? Join our office hours and ask us anything about the new Tinybird. 🗓️ Add it to your Google Calendar: tbrd.co/fwd-ama

Tinybird Forward just launched. DX is 🔥. How about a hackathon? The winner gets over $3,000 in credits from Vercel, Tinybird, Resend, Dub, Plain, Blacksmith, Estuary, Knock, Trigger, LocalStack, and Bucket. But the rules are a bit... interesting... 👀👇

How to build a first version of your analytics backend in a few minutes: Use tb create. tb create is how you get 80% of the way to a production-ready analytics API in about 1% of the time it would otherwise take. youtu.be/TFS-OO63OT4

How to build a first version of your analytics backend in a few minutes: Use tb create. tb create is how you get 80% of the way to a production-ready analytics API in about 1% of the time it would otherwise take.

How to build a first version of your analytics backend in a few minutes: Use tb create. tb create is how you get 80% of the way to a production-ready analytics API in about 1% of the time it would otherwise take.

How to build a first version of your analytics backend in a few minutes: Use tb create. tb create is how you get 80% of the way to a production-ready analytics API in about 1% of the time it would otherwise take.

Tinybird Forward is live on Product Hunt! Help us win the (friendly) Battle Royal against @bucket.co and Novu. Upvote here: tbrd.co/fwd-ph

5 🤯 things you can do with Tinybird Local...

Don't lie. You've run a production service on your laptop before. Learn more about why and how we built Tinybird Local in our latest engineering blog post -> tbrd.co/tb-local

Tinybird is local first. 👨‍💻 You write software code and test it locally before deploying it. Why should data infrastructure be any different? Use Tinybird Local to build, test, and validate your APIs e2e on your machine. tbrd.co/fwd-ama

Hey! We're launching Tinybird Forward on Product Hunt on Tuesday. We're duking it out with @bucket.co and Novu. Most upvotes wins, and we need your help to secure victory. Go here to get notified when we launch. 👇 tbrd.co/fwd-ph

Have questions about Tinybird Forward? Ask us anything. Join our Office Hours next Friday. Here's a Google Cal Link -> tbrd.co/fwd-ama (you don't have to register or give us your email).

🌅 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

Developer laptops are powerful. In the original Tinybird UX, we focused on the cloud and only designed a small part of the platform to work locally. We've received a lot of feedback on this, and decided to revisit it. On Friday March 14th, we're going all on on local dev. Stay tuned.

We've run hundreds of load tests on systems processing petabytes of data in real-time. Here's what we've learned 🧵

Friday changelog highlights: 🔄 Improved S3 connector UI sync status 📊 New Vercel Log Drains template ⚠️ Timeouts moving to 429 status code 🚫 BigQuery deprecation reminder 🐛 Fixed pending queries & UI layout issues More details -> tbrd.co/changelog Have a great weekend.

ClickHouse 25.2 is here. We contributed an important PR to create a new database engine: Backup.

Strategies for running a hybrid OLTP + OLAP stack 👇

So Cursor and Windsurf have added support for MCP Servers. Have you built one? Do you know who is using it? Add remote logging to your MCP server with this open source template 👇 git.new/mcp-analytics

This is the right pattern to follow 👇

So... you boostrapped a simple Datadog alternative. Now scale it.

🚀 Introducing the Logs Explorer, an open source Next.js + Tinybird template to build hyperscale observability features – like Datadog, Vercel Observability, or Supabase Query Logs – insanely fast. Here's why you should deploy it today 🧵 https://youtu.be/JKUakpvQXTg