AI Engineer · LLM Systems · Developer Tooling
I'm Tung, an AI Engineer interested in how systems behave beneath the abstraction layer.
I work on LLM workflows and the backend systems around them. I enjoy reading unfamiliar codebases, tracing unexpected behavior to its root cause, and making changes that are easy to understand and maintain.
Open source is one way I explore those interests, learn from other engineers, and contribute fixes upstream.
| Area | Focus |
|---|---|
| LLM systems | Retrieval, agent workflows, and evaluation |
| Backend engineering | APIs, integrations, and runtime behavior |
| Developer tooling | Testing, debugging, and package management |
A few problems I've worked through in open-source projects:
| Project | Change |
|---|---|
| pnpm | Added dependency validation to the Rust CLI so installs report malformed specifiers instead of silently skipping them. |
| Django Ninja | Fixed async authentication callbacks running twice on synchronous endpoints. |
| jc | Fixed an HTTP header parser crash on whitespace-only lines. |
| versitygw | Aligned Object Lock XML responses with AWS S3 when no default retention is configured. |
I've written up the debugging process behind several of these in oss-case-studies.
Contribution history
111 merged pull requests across 32 upstream repositories (all-time).
Browse contributions · Counts refreshed automatically from GitHub.
Interested in LLM systems, developer tooling, or a tricky bug? Let's talk.