AI Engineer · LLM Agents — Retrieval — Local Inference
M.S. Computer Science, NYU '26 · New York, NY
English · 简体中文
I build systems where the language model is one component among many — retrieval, evaluation
harnesses, GPU data paths, sandboxed tool execution. Most of what I ship runs locally:
on-device inference, no data leaving the machine.
🔍 Open to Fall 2026 internships and new-grad roles starting Jan 2027.
Loomscreen · Wallpaper Engine scenes on macOS
An independent Metal renderer for Wallpaper Engine scenes — not affiliated with Wallpaper Engine.
Also plays video and sandboxed web-page wallpapers, with each connected display running its own source.
Menu-bar first, with playlists, scheduled rotation, and auto-pause on full-screen, game windows,
occlusion, and battery. No accounts, no telemetry. Lite and Pro are split at compile time
(#if LITE_BUILD) rather than by hiding UI.
Analyst-Copilot · local-first agent platform for data analysis
Deterministic profiling and validation come first; LLM workflows extend those facts under constraints. Every claim in a generated report is backed by a claim ledger and checked by hard validators, and open-ended Python analysis runs isolated in a sandbox. Pydantic-typed artifacts end to end, DuckDB for multi-table join discovery and validation.
Urban-Dossier · fully-offline NYC neighborhood intelligence
Click anywhere on a map of New York and get a data-driven dossier on safety, transit, amenities, and building conditions — scored against the city, trended over time, narrated by an on-device LLM, with no cloud calls. One vLLM stack co-tenants Nemotron-30B (NVFP4) and Qwen3-Embedding-4B on a single DGX Spark GB10 alongside a ~30 GiB cuDF/cuVS dataset cache, while cuML DBSCAN clusters incident hotspots across 17 NYC open-data sources.
Clinical-Simulator · AI clinical training platform
Two training modules for medical students. An AI patient you interview from first question to final diagnosis, scored on a 6-dimension rubric with live coverage tracking. And camera-based CPR practice — MediaPipe pose tracking at 33 landmarks / 30 fps, compression detection from wrist-motion peaks, scored on 7 dimensions against the full 30:2 BLS protocol.
rag-agent · hybrid-retrieval agent for vehicle QA
I owned hybrid retrieval — BGE-M3 dense + sparse over Milvus, SQLite FTS, and web search — and a 5-metric evaluation harness covering tool recall/precision, parallel-call correctness, keyword hit rate, and hallucination, which regression-tests every change.
AI Engineering Intern — China-ASEAN Information Harbor · Jan – Mar 2026
Shipped production Dify workflows extracting structured fields from customs and certificate documents (image + PDF), and built an internal RAG assistant — intent recognition, multimodal extraction, tool routing — over a local corpus with zero external data egress.
| Languages | |
| AI / ML | |
| Inference / GPU | |
| Backend / Data | |
| Hardware |



