Fetching metadata from the HF Docker repository... 
Update README.md
29999bf verified - .github Phase 2/3: Gradio Server backend, CRT frontend, engine, agents, mentor, tests, CI/CD
- config Phase 1: project setup, MiniMax-M3 synthetic dataset generation, 1446-row clean dataset
- schemas Phase 1: project setup, MiniMax-M3 synthetic dataset generation, 1446-row clean dataset
- scripts Add Unsloth Modal training pipeline for Nemotron 3 Nano 4B; dataset uploaded
- static Fix AI insight leaking thinking content + make insights on-demand instead of auto-generated
- tests fix: restore llama.cpp with source build, use fine-tuned GGUF model
- training Fix Modal training: use SFTConfig to avoid pickling error
- 426 Bytes Add Modal GPU inference support for faster LLM responses
- 559 Bytes Clean repo, update README, improve .gitignore for hackathon
- 735 Bytes Clean repo, update README, improve .gitignore for hackathon
- 275 Bytes UX overhaul + Gradio migration + remove llama.cpp
- 1.07 kB Add MIT license
- 4.77 kB Update README.md
- 21.4 kB Fix AI insight leaking thinking content + make insights on-demand instead of auto-generated
- 3.77 kB Fix CSS: serve self-contained HTML via ASGI wrapper to bypass Gradio
- 332 Bytes Space: prebuilt llama-cpp-python wheel, minimal Dockerfile
- 2.58 kB fix: eagerly load LLM at startup so /api/health surfaces the real error
- 9.35 kB feat: browser-local engine, Zerodha dashboard, historical events, chatbot, per-user isolation
- 8.89 kB feat: browser-local engine, Zerodha dashboard, historical events, chatbot, per-user isolation
- 4.94 kB Fix AI output: aggressively strip thinking tags, markdown, field prefixes from all LLM responses
- 3.88 kB Add Modal GPU inference support for faster LLM responses
- 24 Bytes Phase 2/3: Gradio Server backend, CRT frontend, engine, agents, mentor, tests, CI/CD
- 219 Bytes Add model download, split requirements, prepare for Modal training
- 158 Bytes Fix: use sdk docker for Python 3.11 compatibility, relax gradio version
