Engineer & Lead - Quanta Manufacturing Fremont - Fremont, CA
(2025-12)
Client - Meta
- Led an engineering team to successful on-site deployment at Meta's Facility in New Albany, Ohio, while developing and modifying automated testing scripts and establishing new testing environments and procedures that streamlined the overall validation pipeline.
- Integrating AI-driven diagnostics into the testing pipeline, improving fault detection accuracy and reducing client-impacting downtime, while building and modernizing testing environments and procedures that increased overall validation efficiency.
AI Engineer - Colaberry - San Jose, CA
(2025-07)
- Built a production multi-agent system (LangChain + custom routing) with a RAG pipeline and vector embeddings. Running GPT-4/Claude as an ensemble, hit 92% accuracy on 50K+ weekly queries
- Cut inference latency from 3.2s to 1.1s with streaming responses, KV-cache optimization, and batched prompt processing on a FastAPI/Kubernetes setup handling 100+ concurrent users
- Shipped a healthcare symptom analyzer and a legal document assistant to 10K+ users. Both use GPT-4/Claude APIs with domain-specific prompt chains and retrieval-augmented generation (35% accuracy lift)
- Set up the AI ops pipeline: embedding drift monitoring, A/B prompt evaluation, model versioning. Degradation incidents dropped 70%, deployments got twice as fast
Co-Founder & Software Engineer - Total Tivio - Remote
(2024-09 - 2025-07)
Co-founded a gamified medical video platform
- Co-founded a gamified medical video platform and built a content recommendation engine using embedding similarity search with Elasticsearch and AI-powered learning-path generation, enabling personalized video suggestions for clinicians
- Added an AI transcription pipeline (Whisper API + AI summarization) with semantic search (vector embeddings + FAISS) so users could find specific surgical procedures across the training library
- Got deployment velocity up 50% with CI/CD (Docker, GitHub Actions); uptime improved 40% with AWS auto-scaling
- Platform and AI codebase acquired by a private buyer. Packaged and transferred everything including prompt libraries, embedding indexes, and inference configs
Technical Mentor - CodeAdvantage - Remote
(2024-12 - 2025-05)
- Instructed 30+ students weekly, developing age-appropriate curriculum and demonstrating complex concepts through clear communication and adaptive teaching methodologies
- Mentored diverse student groups in problem-solving and analytical thinking, showcasing ability to break down technical concepts for varied audiences
Software Engineer, Full Stack - Optum (UnitedHealth Group - Fortune 3) - Remote
(2022-07 - 2023-08)
- Increased processing throughput by 35% by engineering real-time, low-latency data pipeline using Spring Boot, Kafka, Hibernate & MySQL.
- Identified & resolved production performance bottlenecks, improving application stability for 10+ critical services.
- Co-led POC for new internal deployment platform (Quarkus), reducing microservice startup time by 70% & memory footprint by 50%.
- Reduced application startup time by 40% by migrating backend services from Spring Boot to Quarkus, lowering resource consumption.
