Machine Learning Engineer · AppleBay Area, CA ·
I build ML that findsthe answer — and proves it.
Production-scale generative AI at Apple — RAG, LLM evaluation, and personalization for millions of customers. MS CS (AI) at Georgia Tech.
Numbers from production, not notebooks.
- Annual sales lift from recommendations
- $80M+
- LLM-as-judge evaluations processed
- 8M+
- Files in production RAG retrieval
- 107K+
- Fewer production issues via eval gates
- 40%
Three systems, millions of customers.
- 01
Generative AI platform
Production-scale RAG system for millions of customers — prompt engineering with context optimization, content-safety frameworks, and multi-model orchestration.
- RAG
- LLMs
- Safety
- Orchestration
107K+files · 98%+ retrieval accuracy2024 — Now - 02
LLM evaluation infrastructure
Distributed LLM-as-judge framework, automated regression pipelines over benchmark datasets, and quality gates with continuous monitoring.
- LLM-as-judge
- Evals
- Regression testing
8M+evaluations · 40% fewer prod issues2024 — Now - 03
Mac Recommendations
Question-based recommendation engine built from the ground up, plus the Top Model system with A/B testing, Kafka analytics, and real-time buyability suppression.
- Recommenders
- A/B testing
- Kafka
$80Mannual sales lift · 95%+ accuracy2023 — 2024
How I think about it
Great models don’t matter until they retrieve the right context, pass the eval, and ship to real people.
Projects with the code attached.
- 01
Fake News Detector
Trains an LSTM on a labeled news corpus to classify articles as real or fake. Covers text cleaning, tokenization, model training, and evaluation — with a full write-up in the case study below.
Personal · 2024 · Python / TensorFlow / Keras / LSTM
- 02
Fake News Detection API
Puts the fake-news LSTM behind a FastAPI service: training-identical preprocessing, API-key auth, an explicit “Uncertain” band instead of forced guesses, and Docker builds that pull versioned model releases.
Personal · 2024 · Python / FastAPI / Docker / TensorFlow
- 03
Stock Portfolio Tracker
Team capstone for CSC 478: a FastAPI + React/TypeScript app with JWT auth, trades validated against live Finnhub quotes, a watchlist, and real-time portfolio valuation.
UIS · CSC 478 capstone · 2024 · FastAPI / SQLAlchemy / React / TypeScript
- 04
Travel Planner
A Streamlit app that turns a trip's dates into a day-by-day itinerary and keeps activities, flights, hotels, and confirmation numbers in one place, backed by a unit-tested SQLite layer.
UIS · CSC 470 · 2024 · Python / Streamlit / SQLite / unittest
- 05
Mystic Academy
An iOS learning app where courses, units, and lessons live in nested Firestore collections and stream into SwiftUI views live, with Firebase Auth, onboarding, and a financial-literacy starter catalog.
Personal · 2023 · Swift / SwiftUI / Firebase / Firestore