# Sohom Pal > i make agents prove what they did. traces in, failing tasks and a regression eval out. sohom pal. AI engineer. Bangalore, India. ai engineer. building something new right now, more on that here soon. until the end of july 2026 i was at pocket, working on probable, a prediction-market agent, after spending april at network school with the pocket team. before that: lossfunk research on beacon, and safety evals for agents that touch solana. now: heads down on something new. ## Read first - [Home](https://sxohom.xyz/): Sohom Pal, AI engineer - [Home (markdown)](https://sxohom.xyz/index.md) - [Full plain text](https://sxohom.xyz/llms-full.txt): every page in one file - [Colophon](https://sxohom.xyz/colophon/) - [Tools](https://sxohom.xyz/tools/): evals, tracing, testing, sims. [md](https://sxohom.xyz/tools.md) - [Internet](https://sxohom.xyz/internet/): topo trail / metro map of people i keep going back to (60 stations). [md](https://sxohom.xyz/internet.md) - [Things you could buy](https://sxohom.xyz/buy/): desk, sleep, software, and a MacBook Pro ## Projects - [pocket / probable](https://sxohom.xyz/projects/pocket-probable/): the proof layer. [md](https://sxohom.xyz/projects/pocket-probable.md) - [beacon](https://sxohom.xyz/projects/beacon/): sycophancy benchmark. [md](https://sxohom.xyz/projects/beacon.md) - [solana agent safety](https://sxohom.xyz/projects/solana-agent-safety/): gauntlet · sim engine · idl agent. [md](https://sxohom.xyz/projects/solana-agent-safety.md) - [openissue](https://sxohom.xyz/projects/openissue/): deterministic trace to incident. [md](https://sxohom.xyz/projects/openissue.md) - [simtest](https://sxohom.xyz/projects/simtest/): fuzz agents in ci. [md](https://sxohom.xyz/projects/simtest.md) ## Contact if you have an agent you can't prove works, [send me a trace](mailto:hi@sxohom.xyz). - Email: hi@sxohom.xyz - X: https://x.com/sxohom - GitHub: https://github.com/sohomx ## Optional - [Sitemap](https://sxohom.xyz/sitemap.xml) - [robots.txt](https://sxohom.xyz/robots.txt)