01
GRN-Bench — Distributed Benchmark Platform
Self-hosted service running multi-seed graph ML experiments as queued jobs. Idempotent writes survive 500 duplicate submissions and forced worker kills.
- Python
- FastAPI
- PostgreSQL
- Redis
- Celery
- Docker
Machine Learning & Computational Biology
Undergraduate Researcher · University of Missouri — Columbia
01 About
I am a junior at the University of Missouri — Columbia pursuing a B.S. and an accelerated M.S. in Computer Science, with minors in Mathematics and Biological Sciences. What pulls me toward this intersection is how directly these methods are changing the way we understand disease and design treatment.
In Dr. Jianlin Cheng's lab I develop methods for gene regulatory network inference — building the multi-source pipelines, quality-control tooling, and evaluation infrastructure that graph neural network models are trained and measured on. I am co-first author on a NeurIPS 2026 submission under review.
Outside the lab I build systems: distributed job platforms, GPU kernels, and retrieval pipelines. I care about the unglamorous layer — reproducibility, data integrity, and benchmarks you can actually trust.
02 Selected Work
01
Self-hosted service running multi-seed graph ML experiments as queued jobs. Idempotent writes survive 500 duplicate submissions and forced worker kills.
02
Hand-written GPU kernels for the sparse message-passing primitives at the core of graph neural networks, cutting the memory traffic and launch overhead that dominate training.
03
End-to-end pipeline ingesting YouTube captions, metadata, and transcripts into per-creator knowledge bases, with similarity search across 4,000 videos per creator.
04
In-memory key-value store built from the socket up. Custom memory management and a hand-rolled TCP layer serve real-time SET/GET under stress testing.
03 Research & Experience
04 Publications
A. Hedge, K. Zook, R. Kim, J. Cheng. “Agent-Based Gene Regulatory Network Inference.” NeurIPS, 2026.
Co-first author. Developed the automated literature-validation agent and the multi-omic validation component integrating chromatin accessibility and transcriptomic data.
K. Zook, J. Cheng. “A Benchmark for Evaluating Graph-Based Gene Regulatory Network Inference.”
First author. Target submission December 2026, with a public dataset release and reproducible evaluation protocols.
05 Education
GPA 3.84 · High Dean's List, all semesters · Mark Twain Scholarship
Aug 2024 – May 2028
In progress
Mathematics & Biological Sciences