Computer Science + Data Science @ UC Berkeley '26 · GPA: CS 3.91 / DS 3.94
Building infrastructure behind AI systems — distributed systems, ML infrastructure, and healthcare engineering.
Previously: Amazon · American Express · UCSF × Stanford Medical Center · IMDEA Networks (IEEE Cloud Summit 2025) Currently: AI infrastructure at an early-stage startup · Undergraduate researcher, Ivry Lab (motor learning)
DAG-Based Workflow Allocation Across Federated Fog Systems · IEEE Cloud Summit 2025 MILP + greedy scheduling heuristic for distributed task allocation under latency and resource constraints.
LLM Evaluation · UCSF × Stanford Gemini-based scoring pipelines, A/B testing of clinical AI outputs, human evaluation integration on a HIPAA-compliant backend.
Motor Learning · Ivry Lab, UC Berkeley State-space modeling and bootstrap resampling on visuomotor adaptation data (Wang et al. 2025).
- Cognova AI — Multi-tenant Voice AI platform · React · TypeScript · Supabase · PostgreSQL · Twilio · ElevenLabs · Stripe · Clover POS
- Surface Hub — AI-powered B2B growth platform · Next.js · PostgreSQL · Claude API · async cron workers · LinkedIn analytics · ~17k LOC
- Dynamic Prompt Optimizer — FastAPI + React eval framework for AI voice agents · Claude-as-judge · Anthropic API
- Needle — Deep learning framework from scratch · autograd · nn modules · optimizers · ResNet training loop
- SmartSight — Android/Kotlin assistive wearable · OpenCV · TensorFlow Lite · DAG-based edge offloading
Languages: Python · Java · C++ · Go · TypeScript · Bash
Frameworks: FastAPI · Spring Boot · React · Next.js · Supabase Edge Functions
Infra: AWS · Docker · Kubernetes · PostgreSQL · Redis · Kafka · Cassandra · Supabase
ML/AI: PyTorch · Hugging Face Transformers · LangGraph · vLLM · Claude API · Gemini · ElevenLabs
Evaluation: LLM-as-a-judge · Gemini eval pipelines · A/B testing · benchmark design · Weights & Biases