Writing
Working notes on technical due diligence, agentic systems, and the engineering questions that decide investments. Written by our principal, drawn from operating experience rather than research.
March 2026
Access, not model quality, is where agentic systems fail first. Why agents quietly inherit too much power, and what it took in practice to take it back.
read the essay →January 2026
International investors keep evaluating emerging market technology companies with diligence frameworks written for Palo Alto, and the questions those frameworks never ask are the ones that decide whether a deal goes bad. The working framework we use, in five parts.
read the essay →December 2025
A circular lands on a Friday and takes effect in thirty days. How fast that document becomes code in production is one of the most predictive facts about a fintech in an African market, and most diligence never asks.
read the essay →