AndhaKaanun
Jurisdiction-aware AI legal reasoning platform using Indian-law retrieval.
A
AndhaKaanun
Case study visual placeholder
Architecture flow
3 layersProblem
Most legal AI tools give broad answers without jurisdictional grounding, explicit opposing arguments, or enough context to refine weak legal claims.
My Role
Designed the RAG flow, legal context framing, dual-perspective output structure, and missing-facts interaction model.
Solution
A RAG pipeline retrieves Indian-law context, passes it into an LLM reasoning flow, and asks follow-up questions when facts are missing.
Stack
Case study
Problem
Most legal AI tools give broad answers without jurisdictional grounding, explicit opposing arguments, or enough context to refine weak legal claims.
Proof signal
Indian law RAG · IPC grounding · Pinecone · Groq/Llama-3.3-70b
My Role
Designed the RAG flow, legal context framing, dual-perspective output structure, and missing-facts interaction model.
Core product work
- Incident intake
- Citation-aware retrieval context
- Dual argument generation
- Missing fact prompts
Solution
A RAG pipeline retrieves Indian-law context, passes it into an LLM reasoning flow, and asks follow-up questions when facts are missing.
Incident descriptions enter a RAG pipeline backed by Pinecone. Retrieved Indian-law context is passed to Groq/Llama-3.3-70b to produce prosecution and defense reasoning, followed by an interactive engine that asks for missing legal facts.
Architecture Highlights
Vector retrieval over Indian-law context
Separate prosecution and defense reasoning paths
Follow-up question loop for missing legal facts
Challenges and Tradeoffs
- Avoiding broad legal claims without enough retrieval context.
- Separating legal reasoning support from legal advice.
Impact / Outcome
- Built a case-study-ready RAG architecture for jurisdiction-aware legal reasoning.
- Improved output usefulness by forcing both sides of the argument to be represented.
Learnings
- RAG systems need domain-specific guardrails and uncertainty handling.
- Opposing-perspective generation can expose weak or missing assumptions.
Keep exploring