Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Gregory Chin, an AI full-stack engineer with 8+ years of hands-on experience building LLM-powered systems, real-time voice/video platforms, and high-scale AI backend infrastructure. I’ve delivered production deployments across healthcare, fintech, and consumer AI—designing RAG pipelines, multi-agent orchestration, and model/embedding systems that balance accuracy, latency, and reliability. I enjoy building end-to-end products that feel seamless to users, whether that’s a real-time voice agent for clinical intake, an autonomous agent workflow that reduces manual operations, or large-scale event systems powering AI-generated experiences. I’m comfortable across the stack (Python, Go, Java/Spring Boot, C#/.NET, and performance-focused Rust), and I take production readiness seriously—security, monitoring, evaluation, and reproducible MLOps are central to how I ship.…I’m Gregory Chin, an AI full-stack engineer with 8+ years of hands-on experience building LLM-powered systems, real-time voice/video platforms, and high-scale AI backend infrastructure. I’ve delivered production deployments across healthcare, fintech, and consumer AI—designing RAG pipelines, multi-agent orchestration, and model/embedding systems that balance accuracy, latency, and reliability. I enjoy building end-to-end products that feel seamless to users, whether that’s a real-time voice agent for clinical intake, an autonomous agent workflow that reduces manual operations, or large-scale event systems powering AI-generated experiences. I’m comfortable across the stack (Python, Go, Java/Spring Boot, C#/.NET, and performance-focused Rust), and I take production readiness seriously—security, monitoring, evaluation, and reproducible MLOps are central to how I ship.

Gregory Chin

AI Engineer, Full Stack Developer, Back-End Developer, +7

5.0 (2 reviews)

I’m Gregory Chin, an AI full-stack engineer with 8+ years of hands-on experience building LLM-powered systems, real-time voice/video platforms, and high-scale AI backend infrastructure. I’ve delivered production deployments across healthcare, fintech, and consumer AI—designing RAG pipelines, multi-agent orchestration, and model/embedding systems that balance accuracy, latency, and reliability. I enjoy building end-to-end products that feel seamless to users, whether that’s a real-time voice agent for clinical intake, an autonomous agent workflow that reduces manual operations, or large-scale event systems powering AI-generated experiences. I’m comfortable across the stack (Python, Go, Java/Spring Boot, C#/.NET, and performance-focused Rust), and I take production readiness seriously—security, monitoring, evaluation, and reproducible MLOps are central to how I ship.…I’m Gregory Chin, an AI full-stack engineer with 8+ years of hands-on experience building LLM-powered systems, real-time voice/video platforms, and high-scale AI backend infrastructure. I’ve delivered production deployments across healthcare, fintech, and consumer AI—designing RAG pipelines, multi-agent orchestration, and model/embedding systems that balance accuracy, latency, and reliability. I enjoy building end-to-end products that feel seamless to users, whether that’s a real-time voice agent for clinical intake, an autonomous agent workflow that reduces manual operations, or large-scale event systems powering AI-generated experiences. I’m comfortable across the stack (Python, Go, Java/Spring Boot, C#/.NET, and performance-focused Rust), and I take production readiness seriously—security, monitoring, evaluation, and reproducible MLOps are central to how I ship.

Available to hire

I’m Gregory Chin, an AI full-stack engineer with 8+ years of hands-on experience building LLM-powered systems, real-time voice/video platforms, and high-scale AI backend infrastructure. I’ve delivered production deployments across healthcare, fintech, and consumer AI—designing RAG pipelines, multi-agent orchestration, and model/embedding systems that balance accuracy, latency, and reliability.

I enjoy building end-to-end products that feel seamless to users, whether that’s a real-time voice agent for clinical intake, an autonomous agent workflow that reduces manual operations, or large-scale event systems powering AI-generated experiences. I’m comfortable across the stack (Python, Go, Java/Spring Boot, C#/.NET, and performance-focused Rust), and I take production readiness seriously—security, monitoring, evaluation, and reproducible MLOps are central to how I ship.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Beginner

Language

English

Fluent

Chinese

Intermediate

Work Experience

Senior AI Software Engineer at Self-Employed

July 1, 2025 - Present

Built voice AI agents for healthcare and SaaS clients using real-time voice/video and ASR/TTS stacks, improving response time 30–40% and achieving a 4.8/5 satisfaction score. Designed multi-stage RAG pipelines for clinical/insurance/knowledge retrieval with reduced latency and improved relevance. Implemented multi-agent orchestration for healthcare intake, insurance pre-auth, CRM/EMS entry, and appointment scheduling using shared memory and guardrails. Fine-tuned open-source LLMs with LoRA/QLoRA and tracked experiments with MLflow and Weights & Biases; deployed solutions via Azure AI Foundry/Azure OpenAI and used hybrid retrieval with Azure AI Search. Built outbound/inbound lead generation agents integrated with CRMs via webhooks/gRPC, plus fintech/marketing automation pipelines using n8n and FastAPI.

Senior Software Engineer at Butterflies AI

December 1, 2023 - June 1, 2025

Led engineering for an AI social network on AWS Bedrock and LangGraph serving millions of daily users. Architected autonomous AI agent pipelines for thousands of concurrent characters, including memory, persona consistency, safety filtering, and LLM-as-judge evaluation to reduce moderation incidents. Built character creation services using Stable Diffusion/ComfyUI and LoRA training to enable rapid persona generation. Designed event-driven microservices for image generation, feed updates, and notifications with sub-second feed latency at millions of events/day. Implemented scalable Node.js and Supabase backend features including row-level security, auth, and rate limiting; also developed Java ingestion services and C# tooling for analytics pipelines.

Software Engineer at Theoria Medical

February 1, 2019 - November 1, 2023

Led a platform team building HIPAA- and SOC2-aligned AI systems processing thousands of patient encounters monthly. Built production clinical document processing pipelines using LLMs and vision APIs, with structured chunking and vector retrieval (pgvector/Pinecone). Developed a real-time AI medical receptionist (Twilio + ASR/TTS + LLM) for inbound calls including identity verification, intent capture, triage scoring, and routing to reduce call-center volume and handle time. Engineered pre-visit pipelines (voice/form ingestion, LLM summarization, SOAP notes) and automated patient-communication workflows. Designed HIPAA-aligned FastAPI microservices with RBAC, PHI/PII isolation, environment segregation, and audit logging; built additional ASP.NET Core integrations and admin tooling for EHR vendors.

Software Engineer at Dayta AI

August 1, 2017 - January 1, 2019

Built real-time IoT data pipelines for a retail intelligence platform ingesting RTSP camera streams and processing 150k+ daily videos. Developed computer-vision-based demographic tracking (OpenCV), anonymized stream processing, and role-based dashboards. Created script-to-video pipelines using FFmpeg and TTS to automate narrated shopper-behavior reports. Implemented cloud-rendered BI report pipelines with AWS MediaConvert and Three.js to reduce delivery time ~60% for global retail clients. Built Ruby on Rails REST APIs and Sidekiq workers powering multi-tenant ingestion and alert delivery for retail dashboards.