Dakota Folmsbee

Dakota Folmsbee

Computational Chemist
Data Scientist
Materials Scientist

Hire Me!

I’m a PhD-trained computational scientist with a background in chemistry, materials science, and machine learning. For over seven years, I’ve worked across academia and industry to develop tools and models that accelerate scientific discovery. My work has ranged from predicting concrete formulations and small molecules to screening drug candidates using large compound libraries.

What I Do

  • Build machine learning models to predict material and molecular properties
  • Develop web-based analytics dashboards for real-time data exploration (e.g. Plotly Dash)
  • Automate R&D workflows with custom data pipelines and HPC tools
  • Translate complex scientific problems into scalable software solutions

I’m comfortable working in Python, R, and SQL, with experience in libraries like Scikit-learn, PyTorch, TensorFolw, Keras, Pandas, RDKit, and more. I’ve deployed tools in HPC and cloud-like environments and enjoy collaborating across interdisciplinary teams.

What I’m Looking For

I’m open to full-time roles, collaborations, or freelance projects in the following areas:

  • Materials informatics and materials discovery
  • Cheminformatics and computational drug discovery
  • Scientific data science and machine learning engineering
  • Research teams applying AI to real-world chemistry and biology problems

Selected Projects

  • Concrete Mix Optimization Dashboard – Built a production-ready ML pipeline and dashboard for analyzing concrete mix designs at Prometheus Materials
  • chemreps – Created an open-source library for generating molecular representations for ML in chemistry
  • Drug Discovery Pipeline - Led virtual screens of over 100,000 compounds and modeled protein-ligand systems during my postdoc
  • Conformer Benchmark - Compared state-of-the-art ML models conformer energy predictions to conventional quantum mechanical approaches
  • ML Benchmark - Evaluated ML performance on potential energy curves and geometry optimization
  • QTDG - Analyzed torsion angle preferences by comparing crystal structures with gas-phase conformers to create a distance geometry coordinate generation method

Let’s Work Together If you’re building something at the intersection of science, data, and innovation, I’d love to hear from you.

Reach out at dfolmsbee@gmail.com or message me on LinkedIn and take a look at my Resume.