Tabeen Raoof — Bay Area · Available from January 2027
Engineer building applied AI systems — and making them work in production.
At Xylo AI Studios, I own the services that bring financial-advisory firms' email, calendar, CRM and meeting data into one pipeline — and built the monitoring that caught a live outage nobody had noticed. Before that: computer-vision research, and three years delivering enterprise software to customers. MS Computer Science (AI/ML) at SFBU, graduating December 2026.
Now — Backend engineering at Xylo AI Studios, the ProofShape capstone, and two preprints in preparation (TruPhoto and a VLM evaluation study). (Last updated Sep 2026)
Featured
All work →Xylo AI Studios
Forward Deployed / Backend Engineer Intern
Own the ingestion and data-extraction service layer for a communications pipeline serving financial-advisory firms.
Recent notes
All notes →How do you alert on something that's supposed to be quiet?
Most monitoring assumes a steady signal, so absence means failure. Event-driven pipelines break that assumption — here's what it took to alert on one without drowning in false positives.
Why our uncertainty measure was wrong
ProofShape's original uncertainty calculation put generative-completion spread in the denominator for observed surface — which quietly declared every observed region infinitely precise.
An evidence-based decision not to fine-tune
When accuracy plateaus, fine-tuning is the default next move. Three independent signals from TruPhoto said not to — and what to check before assuming more training will help.