Researcher and builder

I study how AI models reason, and build software from what I find

Get in touchSee the code
A · IF WORDING IS ALL THERE ISB · IF THERE IS A BELIEFOne buying question, asked 15 waysThe same question, asked 15 waysFour different brands, no signalOne brand keeps winning, a beliefeach dot = one answer · color = brand namedBelief StabilityWhether AI models recommend the same brand when a buying question is asked in different wordsFollow the research →STORES THAT STOPPED AN AI AGENT BEFORE PAYMENTTop 50k sites, Shopify0 of 17Top 50k sites, other63 of 8850k to 200k, Shopify4 of 6450k to 200k, other16 of 29First 250 stores, Claude · October 2026 · traffic rank from the Tranco top-million listAgentic CheckoutWhether AI shopping agents can complete a purchase at real US online storesFollow the research →1 · WHAT THE AI GETS2 · HIDDEN FROM THE AIComplaint filedNames redactedAI'S VIEW STOPS HEREMotion filedJudge rules3 · THE AI ANSWERSWhich motion? Which lead argument?SCORED AGAINSTThe argument the judge acceptedLegal ReasoningWhether AI models can read the complaint in a real lawsuit and pick the argument that wonFollow the research →

What I'm currently working on

Researching
Whether AI models hold stable beliefs about the brands they recommend, whether they can pick the winning legal strategy from an opening case file, and whether AI shopping agents can buy from 500 US online stores. Read the studies →
Studying
Pursuing a master's in computer science with an emphasis on interpretability and alignment.

Before pursuing research, I was a product marketer in healthcare, helping build GTM strategy for products including Ambient AI, Agentic RCM (billing), and Voice Agents. View my marketing portfolio →