Fauzi Hussein.
I am an Applied AI/ML Lead at JPMorgan Chase. My team builds the GenAI applications used by 230,000 people[1] across the bank, part of a platform that won American Banker's 2025 Innovation of the Year[2]. Before this, I spent eight years at Goldman Sachs, mostly in core engineering and data platforms.
Outside work I build my own tools[3]: a harness that checks the code AI agents write, an implementation of recursive language models, and a few smaller things. When one of them teaches me something, I write about it.
Experience2016 - NOW
JPMorgan Chase
At JPMorgan Chase, I lead applied AI/ML work across LLM Suite, including a PDF GenAI chat application used by 230,000 employees, a multi-agent presentation generation system built on Google ADK, and token-efficiency work that cut LLM costs by 35%. LLM Suite won American Banker's 2025 Innovation of the Year, and I am a named inventor on US Patent Application 19/333,493.
Goldman Sachs
At Goldman Sachs, I spent eight years building core engineering and data platforms. I owned AWS infrastructure for EngHub, a developer documentation platform serving thousands of engineers at 99.9% availability; architected a GPT-4 documentation review system measuring quality across 500 pages; and led a five-person team on a semantic testing framework adopted across departments. The EngHub team received an Engineering Tenet award.