🔥Telegram İfşa

We're stoked to support Nik and the SuperDial team goes through the tough sledding of solving real problems with AI. If you've ever tried to talk to your insurer you know how difficult they make it. Crazy how good this has gotten!

Moving forward, we'll share our release notes on social media in addition to GitHub!

It is very cool to see folks we've never met finding the project and grokking it deeply!

One of the cool things about OSS is that you find out about major deployments after they happen -- exciting to see TensorZero in prod at large organizations!

.Elon Musk your support folks are gaslighting us xAI API regression broke many OSS tools overnight we have thousands of successful CI runs with this behavior — public in our repo

Great work led by aj investigating the limits of automated prompt engineering today! Presumably future AI systems will be able to update based on feedback and editing the prompt is a key primitive for this. Try it out on your problems and let us know how it does!

I'm a huge fan of OpenAI, but this is bad. Evals that lock you into a single LLM provider defeat the entire purpose of an evaluation platform. In a world where LLM benchmarks and leaderboards are being gamed and increasingly meaningless, running your own evals (across models) on

TensorZero is explicitly NOT an agent framework. You can bolt it on to an existing agent and get lots of benefits out of the box! Watch this space…

Shortly after DQN we found tricks to make it work robustly for games. (arxiv.org/abs/1710.02298) I think this paper sets up the pieces for a huge variety of economically valuable tasks and the “rainbow” of tricks that make this happen!

👀

🫶

Sanity check on AI coding agent progress: how many AI-generated PRs are OSS repositories accepting?

really excited for Dynamic Evaluations - a key milestone for some of the crazy RL features we have planned

I switched from a decade of vim to Cursor. Now we can peek at how it works 🕵️. check out how we got under the hood of this bad boy

🫶

It’s been a blast collaborating with Andrew over the past couple months and we couldn’t be more excited to have him on board. We’re excited for our developer community to benefit from the stuff he’s cooking!

TensorZero #1 globally on GitHub this week 🤯 thank you to everyone especially the OSS homies

🥳