Ways to work together.
Three ways to bring me in, matched to where you are.
- Full-timeopen to the right role
Join your team as an AI Product Engineer
Embedded on the team, shipping LLM products from idea to production — with the full-stack rigor to keep them working once real users arrive.
- Contractlimited project slots
Ship a 0→1 LLM product on a fixed engagement
A scoped build, idea to production, by an engineer who has taken four products 0→1. One clear deliverable, one clear timeline.
- Advisory
Advise on agentic / AI-native architecture
Reliability layer, anti-drift, evals, and workflow design — the parts that decide whether an agentic system is trustworthy enough to ship.
Working with Sangmin
Is Sangmin available for hire?
Yes — he's open to AI Product Engineer and Forward-Deployed Engineer roles, US or remote. The fastest way to reach him is sangmlee23@gmail.com.
What does Sangmin build?
He takes LLM products from idea to production — four 0→1 builds — and builds the reliability layer that makes them hold up: retrieval, evaluation, and anti-drift engineering.
What has he shipped?
RankLabs (an AI-visibility engine with a proprietary crawler, 2–4× AI mention coverage), Mindcraftor (a generative-AI SaaS, 1,600+ users), OpenSquid (an anti-drift framework for AI agents), and RecoverPay (full-stack payments recovery).
How long has he been engineering?
Since 1996, at age 12 — the engineering foundation isn't a pivot. His career spans compliance-heavy federal research as a Senior Research Scientist at Peraton Labs and publisher-scale frontend at Complex Networks.
How can we work together?
Three ways: full-time, contract, or advisory. Pick the engagement mode that fits below, or email sangmlee23@gmail.com to start a conversation.
Where is he based, and is he bilingual?
He's US-based and works US hours or remote, and he's fluent in English and Korean.
Building something that has to hold up?
I take LLM products from idea to production, and build the reliability layer that keeps them working — the same anti-drift engineering behind OpenSquid. One engineer, idea to ship.
engineer to engineer — no recruiters, no decks