Stanley.
AI Software Engineer at Garmin Ltd., leading large-scale computer vision and GenAI platforms. I build LLM/VLM systems, design MLOps pipelines, and ship AI from research into production.
Hi, my name is
About Me
I am an AI Software Engineer II at Garmin Ltd. (Nov 2023 – present, promoted Jun 2026), where I lead the development of an organization-wide computer vision platform scaling across 200+ active production services.
I drive GenAI adoption from research to production: by platformizing model access, the GenAI API platform I administer now powers 30+ LLM/VLM use cases across the company. On top of it, I am building an internal agentic AI system that orchestrates the platform’s APIs through LangGraph behind a conversational frontend. I also mentor interns on AI system design and oversee cross-functional AI infrastructure end-to-end.
Previously, I completed my Master’s in Electrical Engineering at NTHU, conducting research in computer vision under Prof. Min Sun, which resulted in a publication at ICRA 2021.
Technologies I work with:- Python
- PyTorch / OpenCV
- Computer Vision
- LLM / VLM
- LangGraph / Agents
- Docker / Kubernetes
- Apache Kafka
- Kong API Gateway
- Grafana
- MySQL / MongoDB
- GitLab CI/CD
- vLLM
- C++ (CUDA/OpenMP/MPI)
Experience
Jun 2026 – present
- Promoted to Engineer II after 2.5 years of leading the organization’s computer vision and GenAI platforms.
- Raise the team’s engineering bar as a code reviewer for the AI team and a technical interviewer for engineering candidates.
- Building an internal agentic AI system: a conversational assistant that uses Garmin’s internal LLM API as its reasoning core and orchestrates APIs from the GenAI platform I administer — spanning language, LLM, and CV categories — as LangGraph tools, so users can complete tasks with text, images, or audio.
AI Software Engineer I
Nov 2023 – Jun 2026
- Led Large-Scale AI Deployment as the primary developer for an organization-wide computer vision platform, scaling and maintaining 200+ active production services.
- Led GenAI Adoption by evaluating 10+ LLM/VLM models through structured proof-of-concepts, producing internal technical assessments that guided company-wide model adoption and shipped 5 use cases into production.
- Mentored 2 interns on AI system design and production deployment, guiding them to independently deliver 3–5 production services and train 5 YOLO models within the AI Surveillance Platform.
Projects
Internal Agentic AI Assistant
LangGraph
LLM Agents
Python
Multimodal
Lead Developer at Garmin (Jun 2026 – present).
Conversational agent built on LangGraph, with Garmin’s internal LLM API as the reasoning core. APIs from the internal GenAI platform I administer — spanning language, LLM, and CV categories — are exposed as callable tools: users drop text, images, or audio into a chat frontend, and the agent plans and chains the right tools to complete the task.
Scalable AI Surveillance & Analytics Platform
Python
Computer Vision
Apache Kafka
MySQL
MongoDB
Lead Developer & PM at Garmin (Nov 2023 – present).
Centralized platform running 200+ camera-stream CV inference services that automate monitoring of physical facility operations. Owns the pipeline end-to-end — stream ingestion, model inference, and real-time anomaly alerts published via Apache Kafka — with detection results persisted to MySQL/MongoDB for analytics.
AI-Powered Document Inspection System
Python
OCR
Computer Vision
Automation
Sole Developer & PM at Garmin (Mar 2024 – present).
Automated document verification combining OCR with CV-based image comparison, cutting manual inspection time by 90% while improving defect detection accuracy. The key design choice: a human-in-the-loop flow that auto-clears low-risk documents, so operators review only flagged anomalies instead of every page.
Enterprise VLM/LLM Research & API Governance
LLM
VLM
vLLM
Kong Gateway
Grafana
AI Researcher & Platform Owner at Garmin (Apr 2024 – present).
Evaluate emerging models and inference frameworks (vLLM and peers) through structured POCs, turning results into technical assessments that drive company-wide model selection. Administer the internal GenAI API platform on Kong Gateway — API key lifecycle, access control, rate limiting — with usage observability via Grafana dashboards. Platformizing model access has scaled adoption from the first 5 production use cases to 30+ LLM/VLM use cases across the company, and now powers the agentic assistant above.
Robust 360-8PA (ICRA 2021)
Computer Vision
Python
Research
Redesigning the normalized 8-point algorithm for 360° FoV images.
Published at IEEE International Conference on Robotics and Automation (ICRA), 2021.
Education
Jul 2020 – Jul 2022
Master's Degree in Electrical Engineering
National Tsing Hua University (NTHU), Hsinchu, Taiwan
Advised by Prof. Min Sun at the Vision Science Lab (VSLab).
Research in computer vision, resulting in a publication at IEEE ICRA 2021:
Sep 2016 – Jun 2020
Bachelor's Degree in Bio-Industrial Mechatronics Engineering
National Chung Hsing University (NCHU), Taichung, Taiwan
Undergraduate researcher at the Precision & Micro Mechatronics Laboratory (PMML), advised by Prof. Yao-Chuan Tsai (Jul 2018 – Jun 2020).
Get in Touch
Open to new opportunities, research collaborations, and technical conversations. Drop me a line.