Responsible AI starts in the IDE, not the boardroom.
25+ years shipping software taught me how frameworks fail at the team level. Now I am watching AI governance make the same mistakes.
Writing from the engineering trenches - Goa, India
Read my writing
Slow thinking in a fast world
I spent 25+ years building and managing software teams across geographies. I have seen frameworks fail not because they were wrong but because they were written for a world that does not match how teams actually work.
AI governance is heading down the same path. I write about the gap between what the policy documents assume and what actually happens when developers point AI at a production codebase.
I write from a farm in Goa, India, where slow living has taught me that the best thinking happens when you stop performing busyness.
Honest writing about AI and the humans building it
The Chicago Sun-Times Printed Books That Don't Exist. Nobody in the Chain Thought to Check.
A freelancer used AI to write a summer reading list. Nine of the fifteen books do not exist. Three organisations were in...
Air Canada Let a Chatbot Make a Promise It Refused to Keep. A Tribunal Said That Was Not Okay.
Air Canada's chatbot gave a grieving passenger incorrect information about bereavement fares. When he asked for a refund...
How We Built a Model Evaluation Harness in a Weekend with Next.js
We built a small tool to evaluate language models for a finance use case. This is what we built, the decisions we made, ...
Before You Trust an AI with Your Money, Test It
I did not know how to pick an AI model for a finance use case. So I tried to build something to find out. I am still not...