As the founding Engagement Manager, I led the build and deployment of a GenAI-powered knowledge platform across three client sites. The system transformed static consulting deliverables into living, queryable intelligence — giving client teams the ability to surface insights from thousands of pages of prior work in seconds.
The Problem
Every consulting engagement produces hundreds of slides, memos, and analyses. The moment the team rolls off, that knowledge becomes inaccessible. Buried in SharePoint folders, forgotten in email threads, or locked in the heads of people who’ve moved on.
What We Built
A retrieval-augmented generation (RAG) system purpose-built for consulting knowledge. Documents are ingested, chunked, embedded, and made queryable through a conversational interface. But the key insight wasn’t the technology — it was the information architecture.
We designed the system around how consultants actually think: by workstream, by hypothesis, by client context. Not by file name or date modified.
Scale
What started as a single-site pilot grew to three deployments across industries. The team scaled from a handful of builders to a 40–50 person cross-regional operation. I transitioned the product from founder-led to a dedicated team of 5+ product managers — the clearest signal that the thing had legs.
Confidentiality Note
This project was built within McKinsey & Company. Specific client names, proprietary architectures, and internal tooling details are confidential. What’s shared here reflects the publicly describable scope of the work.