Add LST-anchored UHI corrector behind UHI_MODEL env var flag

Cut XGBoost/LSTM heat predictor from pipeline and demo

Refresh Neon connection at the start of each pipeline run

Prewarm GraphCast at FastAPI startup to avoid cold-start fallback

Fix pipeline failures: api.py import, NASA POWER resilience, CPU torch

Harden serverless DB checks, CORS, cursor lifecycle, and healing JSON parse

Chronos-Bolt encoder, retrained pricer, 14-day pipeline, sync DB

Rewrite pipeline + DB to fully sync psycopg2 (matches Weather AI 2)

Skip DB writes in pipeline — psycopg2 blocks async event loop

Add flush=True print statements for pipeline logging

Fix pipeline: non-blocking DB startup, fix date.month crash in downscale

Remove blocking reconnect before pipeline run

Restructure pipeline: rule-based heal, template alerts, one Claude review

Fix Neon DB: reconnect before each pipeline run

Re-enable Claude healing when API key is set

Revert Vercel dashboard tracker, add HF Space HTML pipeline tracker

Default to rule-based healing (free) instead of Claude ($3/run)

Parallelize Claude healing (5 concurrent), add pipeline step tracker

Add regionConfig for geographic portability, REBUILD.md fork guide

Polish frontend identity, rewrite tour, add neural pricer eval suite

Fix critical pipeline bugs, purge flood-era code, add ERA5 to Dockerfile

Add pipeline trigger endpoint + weekly scheduler for HF Spaces

Fix issues from /simplify whole-project review

Redesign Program Designer as Coverage Recommendation page

Add neural actuarial pricing engine trained on real Dar es Salaam data

Retrain models on real NASA POWER data, wire pipeline to Neon, enable Claude

Replace React SPA with lightweight API status page on HF Space

Add LSTM neural model, ERA5 data, FAISS+BM25 RAG, Neon DB, eval suite; de-jargon frontend

Initial commit: Extreme Heat Risk Engine