IntentIQ
A two-stage AI pipeline that triages and routes 14K+ emails a day into seven title-insurance workflow categories, born from an order-volume surge that overwhelmed the ops team.
Jay Ozer · AI Engineering · San Francisco
I build production AI for industries where getting it right actually matters: document intelligence, autonomous agents, and whatever a hard problem turns out to need. Mostly I'm interested in turning messy, manual work into something dependable, and in the people I get to build it with.

Selected work
A two-stage AI pipeline that triages and routes 14K+ emails a day into seven title-insurance workflow categories, born from an order-volume surge that overwhelmed the ops team.
HIPAA-ready SOAP-note transcription for dental practices, with a Curve integration that pulls the day's patients and writes notes back automatically. Built so my wife stops staying late.
Fine-tuned GPT-4o-mini chatbot on 750+ verified Q&A pairs, the first line of defense for worried parents on on-call weekends.
Browser-based media optimizer. FFmpeg.wasm compresses video entirely client-side. Your files never leave the tab.
A calm, offline-first PWA for tracking fetal movement, with no accounts, no ads, no tracking. The design DNA behind this very site.
A local-first nursery ledger iPhone app for feeding, sleep, diapers and private voice logging, with optional Supabase sync and a small Next.js marketing/API host.
Experience
Led the applied-AI pod delivering production agentic systems for title-insurance automation, partnering with Fannie Mae and Blend.
Built and led a team of six, establishing engineering rituals and modern data practices that scaled company-wide.
Established a unified source of truth and drove company-wide adoption of modern analytics ahead of IPO.
Led data engineering and BI infrastructure processing terabytes of economic data for monetary-policy decisions in a highly regulated central-banking environment.
Patents
Each one started as a stubborn, manual workflow and turned into something that mostly runs itself. The filings are nice; the part I like is that they quietly save people a lot of time.
LLM agents plus neural search with dual-team extraction, cross-validation, and human-in-the-loop review, replacing a brittle, manual deed workflow with a system that learns continuously.
Hybrid automation across fragmented county portals, with domain-specific parsing, template-driven scripted browsing, AI fallback, and vision verification that every document matches before delivery.
An automated document retrieval system driven by autonomous agents that plan, navigate, and verify on their own, extending the IQ family beyond county records to any source of record. Details coming soon
Working with people
A decade in, the part I care about most isn't the systems. It's the people I build them with. I've hired engineers from zero, sat through the boring reviews, and learned that patience ships more than cleverness. No heroics, just steady, honest work.
Open source & local tools
Python-first coding-agent Kanban that turns a large goal into scoped stories, runs one worker at a time, and keeps an auditable review trail.
View repoLocal-export-first recorder for county sites, preserving full-tab video, DOM evidence, downloads metadata and handoff packs for county-lens.
View repoVoice Presenter turns source evidence from repos, demos, transcripts and notes into a grounded spoken product presentation with live Q&A.
View repoFixed a critical duplicate-download bug, hardening download handling across CDP, JS-fetch and polling paths.
A marketplace of installable AI-agent skill kits, including workflow orchestration, YouTube automation and video understanding tools.
View repoProduction read and write MCP server for OpenAI's AgentBuilder, with dual Graph plus org-account auth. Powers IntentIQ.
Native iOS app where a parent can record a voice, generate a cozy story, and play it back later with cached audio and a local story library.
View repoEdit-before-upload document prep for Voiceflow RAG. The community adopted it; Voiceflow shipped the idea natively.
Ultra-fast LLaMA Guard guardrails for Voiceflow via Groq, real-time content safety with minimal latency.
14 standardized tools exposing my own bio, experience and CV over the Model Context Protocol. Built in under an hour.
A transparent, PRD-driven Bash loop for fresh-context coding iterations across Claude Code, OpenAI Codex and Kimi CLI.
View repoTurns a snack photo into a 4-panel comic where a Gen-Z tooth mascot roasts its threat to your smile. Gemini plus Nano-Banana.
Ask Jay
A retrieval-augmented agent grounded in my notes, projects and patents. Ask how a system is put together, how I think about a hard problem, or just say hi. It answers with citations.
Education & credentials

About
I grew up in Istanbul, played competitive sports, and came up through engineering before finding my way into data and AI. Sport taught me what I still lead by: you rarely win by being exceptional at one thing. You win by being above average at many, and combining them well.
Home is San Francisco, with my wife Andrea, our daughter Chloe, and our dog Hawley. When I'm not shipping, you'll find me at the gym or out on a long walk.



Say hello
I love comparing notes with people building thoughtful AI, a hard problem, a half-formed idea, or just a good conversation. If you're making something interesting, I'd genuinely like to hear about it.