Forward Deployed Engineer / Independent Builder
I’m Larry Xue. I enter ambiguous situations, find the constraint that matters, and build across product, software, AI, infrastructure, and operations until the result works in its real environment.
Field discovery
to production
Selected fieldwork
Public products, private operational systems, and open-source interventions all reveal the same pattern: understand the situation, cross the necessary boundaries, and stay close to the outcome.
01
2026 / sanitized work sample
A sanitized reconstruction of cloud infrastructure console, AI gateway, log search, async task, and internal operations work where dense configuration needs safe, inspectable UX.
02
2026 / public product
A short-form video subtitle generator for creators: upload a clip, generate editable captions, preview creator-style templates, and pay once only when exporting a clean no-watermark MP4.
03
2026 / public product and open-source companion repos
A browser-based product that turns videos, screen recordings, lectures, webinars, and meeting recordings into editable PowerPoint decks, PDFs, image frames, and subtitles without uploading local files.
04
2026 / public fork
A focused OpenClaw contribution fork used to reproduce and patch a Feishu/Lark websocket proxy routing bug, then hand the proof back upstream.
LX / Forward deployed practice
Enter the field
Start with the situation, not a feature list.Talk to the people doing the work, inspect the current system, and find the constraint that actually controls the outcome.
Shape the problem
Turn ambiguity into a tractable deployment.Scope the smallest complete path, make tradeoffs explicit, and choose the architecture after the operating reality is understood.
Build across boundaries
Use whatever layer the problem requires.Move between product, interface, application code, AI systems, infrastructure, and operations without turning the handoffs into blockers.
Deploy and learn
Stay through adoption and feed the field back into the product.Ship into the real environment, observe what changes, remove the next blocker, and turn repeated lessons into reusable product or tooling.
Seen in the work
“Our whole site runs on Larry’s Astro Sassify template. He didn’t just hand it over. He gave us practical, down-to-earth advice on getting it launched. Genuinely went above and beyond.”
Field notes
Notes on product decisions, systems, AI, interfaces, failures, and the lessons worth carrying into the next deployment.