Geo Regulation

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Geo Regulation

RAG with LlamaIndex to pull the right law snippets for each feature Reasoning LLM generates the classification + explanation using only retrieved context Voting/Consensus using model diversity: review and vote Strict vs risk-oriented Fine-tuned small model (HF) acts as one of the voters for speed/cost and domain alignment Evidence trail (snippets used, model IDs, JSON outputs) saved for audit

Built With

• csv-logs-tools:-markitdown-for-pdf?md
• frontend:-streamlit-(demo-ui)-backend:-python-+-llamaindex-(rag)-models:-huggingface-(baai/bge-embeddings
• json-catalog
• pydantic-schemas
• qwen/phi-for-llm)-storage:-faiss/chroma-vectordb