Thoughts on MDM, data architecture, AI product development, and building a consulting studio from the ground up. Written by Babacar GUEYE.
When two companies merge, their data landscapes collide. Here's what I learned designing the MDM strategy for a major acquisition at Givaudan — referential conflicts, rebinding waves, and what actually breaks.
After 10 years building MDM platforms across banking, insurance, and manufacturing, I keep seeing the same mistakes. MDM is not a software project. It's a governance program that happens to need software.
How I'm using FastAPI, pgvector, and Gemini to build an invoicing assistant for small businesses in Senegal — and what I learned about the gap between LLM demos and production-grade products.
Running an Anti-Money Laundering program under ACPR pressure, coordinating four countries, and shipping daily screening at scale. What the technical side of financial compliance actually looks like from the inside.
I've been a consultant for 10 years. Now I'm building Bakene — a consultancy and product studio. This is what the transition looks like: the legal setup, the positioning, and the uncomfortable parts nobody talks about.
The skills that make a good enterprise architect — systems thinking, data modeling, IT/Business translation — turn out to be exactly what you need to ship AI products that actually work in production.