Skip to content

Tung Nguyen / Independent AI engineer

Make your AI
workflow easier
to run.

From unclear processes to agent harnesses and operating guardrails. Practical engineering, starting with one workflow.

Services

Where is the workflow getting stuck?

Three ways I help teams move from brittle demos to systems they can run.

Featured work

Saved-records evidence from the Engineering the Agent Loop recording Recorded showcase

Engineering the Agent Loop

A custom harness with host-owned completion.

20 records validated, saved, and independently verified from one fixed test page.

Watch the showcase

Edited walkthrough of a retained run. Broader generalization is future work.

Next step

Bring one workflow.
Find a practical next step.

About the practice

Close to the work.
Clear about the limits.

The work stays close to the point where a model’s answer becomes someone else’s work. That means tracing the real workflow, making outputs explicit, testing the edges, and designing a useful handoff when the system is uncertain.

The goal is not autonomy for its own sake. It is a system your team can operate, question, and improve without pretending the hard cases disappeared.

Explore workflow examples