Experience
Where I've Worked
Building real systems on real teams, from silicon validation to AI tooling.
AMD (Advanced Micro Devices)
Software Engineer Intern
Jan 2026 – Aug 2026
Austin, TX
- Building Python feature-validation tooling for CPU power management on AMD's next-generation EPYC server platform (Venice), enabling repeatable regression testing across 10+ silicon features
- Built dual-socket, multi-IOD support from scratch, expanding the team's validation coverage across system topologies
- Built AI-powered tools that auto-generate test configurations from specifications, reducing manual setup for a team spanning the US, Malaysia, and India
Arkane Digital
Software Engineer Intern
Jun 2025 – Aug 2025
Remote
- Developed and deployed 10+ AI-powered workflows on the Raikoo platform to automate the front-end design-to-code pipeline
- Built multi-component generation logic with text-preprocessing pipelines, supporting 20+ UI configurations across internal product teams
Real-world Results
Featured Projects
See how I transformed concepts into engaging digital experiences.
Personal Project
DocAgent
- Full-stack RAG system on async FastAPI + Postgres/pgvector with streaming answers, source citations, and validated agentic tool use
- 239 tests running against real Postgres in CI, with a 50-question eval suite gating PRs on quality regression (93% answer correctness)
- Dockerized and deployed as a public live demo on Hugging Face Spaces with sub-500ms first-token latency
Personal Project
Financial Transaction Engine
- Immutable double-entry ledger engine in Go with idempotent transfers/reversals and authorize→capture→release fund holds, race-detector-clean under concurrent load
- Found, root-caused, and fixed a real concurrency bug under -race, documented as a full incident-style write-up with root-cause analysis
- Chaos-tested for crash safety (SIGKILL mid-transaction) with zero lost or duplicated money, backed by Prometheus RED metrics and a provisioned Grafana dashboard
Personal Project
TelemetryBench
- Containerized full-stack app (FastAPI + Streamlit + SQLAlchemy/SQLite) that ingests raw DAQ CSVs from hardware parameter sweeps
- 8 REST endpoints for sweep ingestion, multi-run comparison, time-series queries, and anomaly scoring
- ML-based run-quality scoring with a scikit-learn Isolation Forest that catches drift cases static thresholds miss
Performant
Responsive
Analytical
Resourceful
Scalable
Efficient
Passionate
Collaborative
Detail-Oriented
Innovative
Versatile
Performant
Responsive
Analytical
Resourceful
Scalable
Efficient
Passionate
Collaborative
Detail-Oriented
Innovative
Versatile
Performant
Responsive
Analytical
Resourceful
Scalable
Efficient
Passionate
Collaborative
Detail-Oriented
Innovative
Versatile
Performant
Responsive
Analytical
Resourceful
Scalable
Efficient
Passionate
Collaborative
Detail-Oriented
Innovative
Versatile
About Me
A Glimpse Into My World
I grew up in Keller, Texas, and I'm now studying Computer Science at Texas A&M, previously interning at AMD in Austin. Outside of code, I'm usually playing basketball, cooking, or reading.
My Reads
Mere Christianity
by C.S. Lewis
My Toolbox
Explore the technologies and tools I use to craft exceptional digital experiences.
Python
FastAPI
Docker
PostgreSQL
scikit-learn
TypeScript
JavaScript
React
Next.js
TailwindCSS
Git
GitHub Actions
Java
C++
C#
Python
FastAPI
Docker
PostgreSQL
scikit-learn
TypeScript
JavaScript
React
Next.js
TailwindCSS
Git
GitHub Actions
Java
C++
C#
Python
FastAPI
Docker
PostgreSQL
scikit-learn
TypeScript
JavaScript
React
Next.js
TailwindCSS
Git
GitHub Actions
Java
C++
C#
Python
FastAPI
Docker
PostgreSQL
scikit-learn
TypeScript
JavaScript
React
Next.js
TailwindCSS
Git
GitHub Actions
Java
C++
C#
Beyond the Code
Explore my interests and hobbies beyond the digital realm.
Basketball🏀
Cooking🍳
Hiking🥾
Gaming🎮
Music🎵
Fitness🏋️♂️
Reading📚
Fishing🎣
Let's Connect!
I'm always interested in discussing new opportunities, collaborating on projects, or just talking about tech. Feel free to reach out at nathanwjchung@gmail.com