Portfolio

Structured profile

LLM-readable profile

A plain-language rendering of the same verified data used on the homepage.

Summary

Chun-Yuan Hsu is an engineer and researcher with a background in civil engineering, computer science, and data science. Master of Science in Data Science student at the University of Michigan building AI agents, backend infrastructure, and applied machine learning systems across production and research environments.

Experience

TSMC — Digital Workflow Development Department Intern

Hsinchu, Taiwan · May 2026 – Present

  • Developed an AI-agent workflow with the Claude Agent SDK to automatically triage backend alerts and generate structured incident-analysis reports.
  • Built integrations to ingest alerts from Alertmanager and retrieve logs and metrics from Kubernetes workloads via ELK and Prometheus.
  • Designed a hypothesis-driven investigation loop that correlates alerts, logs, and metrics to identify likely root causes and summarize actionable findings for engineering teams.

QNAP — Backend R&D Internship

Taiwan · Jan 2025 – Jul 2025

  • Built a retrieval-augmented Jira issue search system using AWS Bedrock and ChromaDB embeddings, increasing developer issue-resolution efficiency by 50%.
  • Developed an MCP-based Jira search server that integrates with IDEs, enabling developers to query and explore issues directly from their coding workflow.
  • Refactored Device Avatar microservices from Python to Go, achieving a 30% performance gain and optimizing deployment on Kubernetes.
  • Diagnosed and patched a critical memory leak in cloud production by correlating Grafana metrics with execution traces.

Lilac

A cross-cloud Infrastructure-as-Code lifting framework that reconstructs Terraform configurations from existing deployments across Azure, Google Cloud, and AWS.

  • Integrated LLMs with symbolic verification to learn resource-dependency mappings from cloud APIs.
  • Evaluated the system on real cloud environments, achieving higher accuracy and coverage than existing tools such as Terraformer while maintaining correctness.

Toward Interpretable Brain Age Prediction and AD Classification

An interpretable pipeline for brain-age regression and Alzheimer’s Disease classification using structural MRI scans from OpenBHB and ADNI.

  • Adapted NeuroVFM, a pretrained 3D vision transformer, to extract patch-level MRI embeddings and aggregate variable-length features with attention-based multiple instance learning.

  • Trained joint classification and regression heads while mapping attention back to clinically relevant brain regions.

  • Reported result: 0.873 diagnostic accuracy
  • Reported result: 0.775 macro F1
  • Reported result: 3.54-year MAE
  • Reported result: 0.966 R²

VizThinker

A graph-based interface for interacting with LLMs that reimagines linear chat as a visual conversation graph.

  • Implemented branching and node-based history navigation to support complex idea exploration.
  • Deployed on Google Cloud Platform with Node.js, React, and Python.

Additional projects

  • Jira Issue Search: A retrieval-augmented Jira issue search system using AWS Bedrock and ChromaDB embeddings.
  • Issue Search MCP: An MCP server that exposes natural-language Jira query, suggestion, and issue-retrieval tools to coding workflows.
  • File Translator: A Gemini-powered tool that translates English PDF documents into Traditional Chinese while preserving layout through generated LaTeX.
  • AZtec Image Comparison: A computer-vision tool for detecting and comparing overlapping patterns in crystallographic pole-figure images.
  • MIPS CPU Architecture: Verilog coursework covering MIPS assembly, an ALU, a single-cycle CPU, and a pipelined CPU with forwarding and stalling.
  • OS Nachos: Operating-systems coursework implementing system calls, multiprogramming, virtual memory, and file systems in Nachos.
  • Advanced Compiler: LLVM coursework implementing data-dependency and pointer-analysis passes and studying array languages.
  • Quantum Event Identification and Simulation of Quantum Event-Learning Procedures: Python simulations comparing quantum random and blended measurements for quantum event identification.

Skills

  • Languages: Python, C++, Go
  • AI & ML: PyTorch, AI Agents, Claude Agent SDK, LLM Integration (AWS Bedrock), RAG (ChromaDB), Embeddings, Semantic Search
  • Cloud & DevOps: Docker, Kubernetes, AWS, GCP, GitLab CI/CD, Prometheus, Elastic Stack (ELK), Alertmanager, Grafana, NATS
  • Frameworks & Systems: Node.js, React, Linux (Debian, Arch), Git, Scrum

Education

  • Master of Science in Data Science, University of Michigan, Ann Arbor, MI, USA — Sep 2025 – Present
  • Master of Science in Computer Science, National Tsing Hua University, Hsinchu, Taiwan — Sep 2022 – Jan 2025
  • Bachelor of Science in Civil Engineering, National Cheng Kung University, Tainan, Taiwan — Sep 2018 – Jun 2022

Personal background

Originally from Tainan, Taiwan, Chun-Yuan moved from civil engineering into computer science after discovering a stronger interest in programming. Outside technical work, his interests include sports, fitness, darts, and Linux ricing.