routsom/causeval

Statistically rigorous, causal evaluation for LLM apps on top of DeepEval: confidence intervals, causal interventions (RAG grounding, perturbations, agent attribution), and judge validity (bias audits, calibration, PPI).

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

Hacker News discussions

• Causeval – statistically rigorous, causal evaluation for LLM apps (1 points, 0 comments, by somrout, 2026-10-02)