# Elvis Saravia urges building evals to measure open-ended AI improvements

> Original title: Learn to build evals, folks!

- Source: Elvis Saravia
- Published: 2026-10-31T16:39:58.000Z
- Bad Influence: https://badinfluence.news/items/r0rn974hyo8pl02c0sgkumc76
- Original: https://x.com/omarsar0/status/2116570890758402382

## Summary

Elvis Saravia says recursive self-improvement requires knowing how to measure what you want to improve, whether a model, codebase, or business function. He argues AI excels mostly at verifiable problems and that open-ended work needs different methods and rigorous evals rather than short-term benchmark chasing. He says DAIR.AI is researching and building evals in this space and invites collaboration.

## Post

Learn to build evals, folks!

You are not RSI-pilled enough if you don't know how to properly measure the thing you want to improve.

It doesn't matter if it's a model, complex codebase, or a business function. This is extremely hard in the real world.

You ever wonder why most of what AI is great at today is around verifiable problems? 

The open-ended world requires a completely different approach to improving and measuring models.

We need to go back to hardcore science and evals. Enough of this short-term benchmaxxing.

This space offers so much opportunity. We are doing research and building evals in this space @dair_ai. I am currently chatting with folks building in this space too. Feel free to reach out for potential collab opportunities or just share perspectives.