praxagent/praxagent-ai

Public research on LLM internals, Jacobian lenses, SAE steering, nonlinear dynamics, evaluation, and inspectable AI systems.

4 best points · 0 comments · 1 Hacker News post · 2 GitHub stars · View on GitHub

Hacker News discussions

• Prax: An agent runtime that learns from past mistakes and fixes code in a loop (4 points, 0 comments, by steveharing1, 2026-04-23)