AI4PFAS
Deep learning workflow and dataset for toxicity prediction.

Cost-aware computing
Automating cost-aware profiling, prediction, and provisioning of cloud and HPC resources.

Clouds
Self-supervised data-driven methods for the application on big satellite clouds imagery

CODAR
We develop methods for online data analysis and reduction on exascale computers

Colmena
Machine Learning-Based Steering of Ensemble Simulations

Data and Learning Hub for Science (DLHub)
A simple way to find, share, publish, and run machine learning models and discover training data for science

Flows
We are developing methods to automate the scientific data lifecycle.

Foundry
An open source machine learning platform for scientists

funcX
funcX is a Function as a Service platform for scientific computing.

Gladier
Globus Automation for Data-Intensive Experimental Research

Scientific Language Modeling and Information Extraction
Data mining from literature with a foundational science-focused language model

KAISA: Scalable Second-Order Deep Neural Network Training
KAISA is a novel distributed framework for training large models with K-FAC at scale.

The Materials Data Facility
We are creating data services to help materials scientists publish and discover data

Parsl
Parallel programming library for Python

ProxyStore
A data fabric for high-performance, multi-site workloads facilited by transparent object proxies.

SZ3 Compression
We develop a series of prediction-based lossy compression algorithms for scientific simulations.

Whole Tale
Whole Tale is a cloud-hosted platform for conducting and sharing reproducible science