AI Infrastructure · LLM/RAG · ML Platforms · Agents
Min Liu
I build reliable AI systems: LLM workflows, RAG products, voice analytics pipelines, ML platform infrastructure, and developer tools. Based in Vancouver, BC, Canada; former Alibaba Senior Algorithm Expert; Master of Engineering, University of Victoria.
Open To
AI infrastructure, applied AI, ML platform, agent systems, and developer tooling roles with teams in Canada or remote-friendly North American organizations.
Technical Strengths
- Applied AI products: LLM workflows, RAG, voice/transcription pipelines, structured outputs, dashboards, and evaluation.
- ML platforms: ranking systems, feature pipelines, model serving, experimentation, and production observability.
- Full-stack AI applications: authentication, Postgres data models, operational workflows, and deployment-ready infrastructure.
- Systems and developer tools: C++, Rust, Python, Clang/LLVM, and practical debugging workflows.
Selected Work
Healthcare AI · RAG · regulated workflows
OncoMate
AI workspace for oncology equipment operations, designed around a real user pain: clinical and service teams lose time when manuals, maintenance history, and repair notes live in different places.
OncoMate treats the assistant as part of an operating workflow, not just a chat box. The product brings technical documentation and service context into one place so a team can move from a vague equipment issue to grounded troubleshooting steps, clearer records, and faster handoffs.
I designed the system around the constraints that matter in regulated operational work: authenticated users, structured equipment records, retrieval grounded in source material, and workflows that help people decide what to do next instead of simply generating text.
Built with Next.js App Router, Supabase Auth/Postgres/Edge Functions, OpenAI APIs, Tailwind, and RAG-oriented data modeling.
Voice analytics · LLM pipelines · product infrastructure
Actual Voice
Voice intelligence platform for teams that need to understand open-ended feedback without manually reading every transcript or exposing raw individual responses.
The user value is not transcription by itself. The useful outcome is turning messy spoken responses into patterns that leaders can act on: engagement signals, recurring themes, and dashboard-ready summaries that preserve the difference between individual raw data and aggregate insight.
This kind of product depends on both product judgment and infrastructure discipline. The pipeline has to handle uploads, transcription, LLM analysis, scoring, aggregation, retries, tenant boundaries, and dashboard contracts while keeping the experience simple for the people reviewing the results.
Built with Supabase Storage/Postgres/Edge Functions, worker queues, Whisper, LLM classification, dashboards, and production pipeline hardening.
Developer tools · C++ · compiler infrastructure
acav
Clang AST viewer for engineers, researchers, and students who need to understand C++ source structure without fighting raw compiler dumps.
Compiler tooling is powerful, but the normal debugging experience is often hostile: huge AST dumps, hard-to-follow source locations, and little help connecting compiler facts back to the code a person is trying to reason about. acav turns that into an inspectable visual workflow.
The product value is faster understanding. It helps people teach compiler concepts, investigate complex C++ constructs, and design static-analysis or refactoring tools with clearer evidence from the compiler layer.
Built with C++, Qt 6/QML, Clang/LLVM AST APIs, source-location indexing, lazy tree rendering, structured metadata extraction, diagnostics, and GitHub Pages documentation.
Who I Am
I have worked across large-scale ML ranking systems, feature pipelines, model serving, AI product engineering, and low-level developer tooling. My background combines production ML platform experience, applied AI product work, and systems-level engineering.
At Alibaba, I led ML platform work across CTR/CVR ranking, causal marketing science, SEO automation, feature pipelines, and production model-serving systems. I am now based in Vancouver, Canada, building portfolio-grade AI products while looking for AI infrastructure, ML platform, applied AI, agent systems, or developer tooling roles.
Capabilities
- AI product engineering: copilots, RAG systems, model-backed workflows, user-facing dashboards, auth, deployment, and evaluation loops.
- Agent and LLM systems: tool-using workflows, retrieval boundaries, fallback behavior, transcript/voice pipelines, structured outputs, and observability.
- ML platform engineering: feature pipelines, ranking systems, continuous training and serving, experiment design, and production model operations.
- Systems and tools: C++, Python, Rust, Clang/LLVM, structured data tooling, and practical infrastructure for developer workflows.
Stack
Python · C++ · Rust · TypeScript · Next.js · Supabase/PostgreSQL · Docker · OpenAI/LLM APIs · RAG · Whisper · Clang/LLVM · ML ranking systems
Contact
I am open to AI infrastructure, applied AI, ML platform, agent systems, and developer tooling roles in Vancouver, across Canada, or with remote-friendly North American teams.