Catie
Engineer · Founder · Patient
Building hormonal health infrastructure for 300M women.
300M+ women globally live with hormonal conditions. The best resource most of them have is a Reddit thread from 2019.
I have endometriosis and PCOS. The most useful things I learned came from other patients online, not from any doctor or app. Which phrases get your GP to actually listen. How to prepare for surgery. Which products might be making your specific conditions worse. That knowledge existed, scattered across forums, never built into anything.
Embyr pulls it together. Hormone-aware product intelligence built around you: your conditions, your symptoms, your cycle and (soon) your bloods. Every recommendation traces back to peer-reviewed research. Multi-condition aware, because PCOS and endo together isn't the same as either alone. GP Prep Kits that turn your data into appointment ammunition. Community wisdom where the best patient knowledge rises to the top.
I'm building the full stack solo. Monorepo with isolated packages for scoring, product lookup, and data access. The scoring engine and product API are shipping code, not slides.
3+ years building production systems for 500+ traders at Morgan Stanley. Now building the infrastructure that should already exist.
What I've Built
pnpm workspaces + Turborepo · TypeScript throughout · Expo React Native + Next.js 15 · Isolated packages with clean APIs
Scoring Engine
Isolated TypeScript package. Ingredient parsing, normalisation with fuzzy matching, and condition-specific scoring against 56 PubMed-cited studies. Deterministic, algorithm-versioned.
OCR & Product Pipeline
Barcode lookup chain with circuit breakers per external API and in-memory TTL cache. Falls back through Open Food Facts, Open Beauty Facts, then OCR. Normalises heterogeneous responses into a single product type.
Community & GP Prep
Upvote-ranked patient knowledge. GP Prep Kits generate appointment packs from scan history with condition-specific talking points and cited evidence.
Privacy-First Architecture
Zero ad SDKs. GDPR consent with policy versioning, hard-delete cascade, row-level security. Health data never leaves the user’s control.
Track Record
Morgan Stanley
Analyst · 2022 – Present
Engineered Java and Python backend services for an intraday PnL and risk platform used by 500+ equity traders. Currently building a provenance-aware NLP pipeline to replace $1M+/yr third-party vendor tooling.
EF Polaris Fellow
1 of 24 selected · 2024
Entrepreneur First’s selective fellowship on how institutions emerge, scale, and fail. Built frameworks on incentives, legitimacy, and coordination. Applied directly to building trust in health-tech.
YC AI Startup School
San Francisco · June 2025
Y Combinator’s programme for high-potential technical founders. Immersed in rapid product iteration and early-stage company building.
50K+ Organic Audience
Instagram · Built from zero
Grew 50,000+ followers organically. That content instinct and community-building ability is now the engine behind founder-led distribution for Embyr.
Software Engineering BSc
Queen Mary, University of London · 2022–2026
Expected First Class Honours. A*A*A at A-Level in Economics, Maths, and Software Development.
National Competitor & Performer
Judo · Gymnastics · HBO · BBC · West End
National titles in judo and gymnastics. Performed on Game of Thrones (HBO), BBC, and West End. Same intensity now behind building Embyr.
Previous Builds
Embyr
See what I'm building · embyr.health