Resume


Hi, I’m Zach Maas 👋

PhD computer scientist working on interpretable AI and computational genomics. I build models we can understand and software that doesn’t waste your time.


TL;DR

FocusPhD ‘25: Bayesian models + transformers for genomics, now doing mech interp on LLMs
Experience6+ yrs Python ML/data stack, HPC infra, and scientific computing
CurrentIndependent research on sparse representations in language models

Current Work

Independent Researcher • May 2025 – Present
Mechanistic interpretability and open-source tooling


Previous Work

RoleTimelineHighlights
LincSwitch Therapeutics
Data Science Consultant
Jun–Dec 2024Built greenfield data infrastructure (Nextflow, HPC, R) for early-stage biotech
University of Colorado Boulder
PhD Researcher
Aug 2020 – May 2025- BERT + SAEs for genomic regulatory grammar
- Bayesian MCMC for RNA-seq uncertainty quantification
- SOTA cyanobacteria segmentation (U-Net)
University of Colorado Boulder
Research Assistant
May 2019 – Aug 2020Large scale SQL-backed ETL system for genomics pipelines

Stack

CategoryTools
ML/DLPyTorch, HuggingFace, scikit-learn, PyMC, CUDA, W&B
DataPandas, NumPy, SQL, Nextflow, Docker
LanguagesPython, Bash/Awk, R, C++, JS/TS, Go, Zig
InfraLinux, SLURM, Terraform, AWS

Selected Publications

YearTitleVenue
2025*Supervised and Unsupervised Methods for Transcriptional Sequencing DataPhD Thesis
2025TFIIH kinase CDK7 drives cell proliferation through core TF networkScience Advances
2024*Internal and External Normalization of Nascent RNA SequencingBMC Bioinformatics
2024*Deconvolution of Nascent Sequencing Data Using TREsPSB
2021Transcription Factor Enrichment AnalysisCommunications Biology

* = first author


Education

DegreeInstitutionNotes
PhD, Computer ScienceCU Boulder (2025)Dissertation: Supervised and Unsupervised Methods for Transcriptional Sequencing Data
B.A., Mathematics & ChemistryCU Boulder (2019)Summa Cum Laude

Misc

Ham radio (AE) • 3D printing • Homelab experiments