Research master's training with an emphasis on ecological data analysis and computational methods.
Research-informed engineering
Search, recommendation, and online intelligence.
I build production-minded retrieval, ranking, and online-serving systems, while keeping a parallel research interest in AI-assisted ecology and coral conservation.
Current target
Algorithm Engineer
Primary build
Qilin Search & Recommendation System
Parallel interest
AI + multimodal coral research PhD
Profile
Engineering first, with research discipline behind it.
Biology foundation that later expanded into transcriptomics, eDNA, and bioinformatics workflows.
Focused on systems that can be trained, deployed, and iterated under real engineering constraints.
Research Thread
Research themes that still shape how I think about systems.
Coral community assessment with eDNA and CoralNet-assisted analysis
Combining ecological workflows with AI-assisted annotation to interpret coral community structure.
Airborne environmental DNA for British bat surveys
Analysis and presentation work around airborne eDNA metabarcoding in applied ecological research.
Deep-sea mussel transcriptomics
Transcriptomic analysis related to host defense, adaptation, and extreme marine environments.
Featured Build
Rednote Qilin Search & Recommendation System
Built a unified search-and-recommendation pipeline from data preprocessing and feature engineering to retrieval, coarse ranking, final ranking, deployment, and online serving.
Combined DSSM dual-tower retrieval with Faiss, Swing, and UserCF; used LambdaMART with LightGBM and XGBoost for coarse ranking; and DIEN for sequence-aware final ranking.
Served through FastAPI and Redis, with real-time behavior writeback and linked search-recommendation updates. End-to-end latency stays around 150ms.
Stack
Core tools and systems.
Support
Buy me a coffee.
If my work is useful to you, you can support it here.