Welcome to my technical portfolio.
Here you can learn about some of the cool things I've built!
Software Engineering
My software engineering journey started with the CHNOSZ R package for thermodynamic calculations and diagrams. I have maintained this package on CRAN since 2009 and it is used by 100+ research groups around the world. One of the reasons for its success is data quality (blog post) as much as the computational and visualization features.
NLP Engineering
My background in scientific software combines with a passion for writing to build robust claim-verification systems in the field of Natural Langauge Processing. I’ve mastered the art of fine-tuning transformer models (blog post). My approach speaks to the MLOps paradigm of iterating on the data to create a powerful evidence-retrieval pipeline. This result is a state-of-the art system that helps writers and editors verify quotation accuracy in scientific publications.
ML Engineering
I wrote the pyvers Python package for reproducible data processing and iterative ML experiments with the pytorch-lightning framework for deep learning. Putting it all together with litserve to deploy the model at scale and a Gradio frontend makes it easy for anybody to use the claim verification system. (Credit for citation example: Sarol et al., 2024.)