BlackRock · 2022 – 2025
BlackRock: GenAI RAG for Legal Fund Verification
90% better reporting accuracy and 100% adoption.
- Role
- Analyst (Product Management)
- Tools
- Llama · Python
- Focus
- GenAI · RAG · Fintech
01Problem
Legal fund verification meant working through 100+ documents, and reporting accuracy depended on getting it right.
02My role
Analyst (Product Management), Risk & Quantitative Analysis. I owned market positioning, user segmentation and the roadmap.
04Decisions & trade-offs
- Chose a Generative AI Retrieval-Augmented Generation (RAG) approach to automate the verification.
- Partnered on fine-tuning a Llama model for domain precision.
- Defined who the tool was for and how it would be positioned, then sequenced the roadmap around them.
05Outcome & metrics
- 90% improvement in reporting accuracy.
- 100% adoption.
