Welcome to my open-source portfolio site, built to highlight technical projects, machine learning pipelines, and scientific tooling at the intersection of materials science, sustainability, and data engineering.
🔗 Live Site: https://acfdavis.github.io
Parsed 1,200+ ThermoML XML files to extract structured thermophysical property data (e.g., thermal conductivity) for machine learning workflows. Includes robust CLI, uncertainty handling, and FAIR-inspired data modeling.
Built a machine learning pipeline using Citrine + UCSB datasets. Applied matminer-based feature engineering, trained ensemble models (e.g., XGBoost, Random Forest), and interpreted results using SHAP.
New Dev Branch:
I’m actively developing a side branch to explore:
-
Deep learning with TensorFlow + PyTorch
-
Dockerized training and deployment
-
ML experiment modularization
Co-author and data lead on peer-reviewed publications in:
- Scientific Reports – CFD modeling of cough aerosol in aircraft cabins
- ASHRAE Journal – Particle transport in buildings
- Journal of Applied Microbiology – SARS-CoV-2 inactivation on cabin materials
- SAMPE Proceedings – Adhesion optimization for thermoplastic composites
I’m a sustainability-focused R&D engineer with 13+ years of experience bridging data science, materials engineering, and technical leadership. I specialize in building reproducible scientific workflows and ML-ready datasets for physical systems.
I care deeply about:
- Open-source science
- Sustainable materials
- Scalable modeling pipelines
- FAIR and interpretable data practices
You can clone this site to build your own data science or engineering portfolio:
git clone https://github.com/acfdavis/acfdavis.github.io.git
cd acfdavis.github.io