Bayesics, LLC

Bayesics, LLC, doing business as Bayesics LLC, is a minority-owned small disadvantaged business specializing in advanced data analytics and modeling services for federal agencies, particularly NASA. Registered to receive federal contract and grant awards since May 2013, Bayesics leverages its expertise in machine learning, Bayesian statistics, and digital twin development to deliver innovative solutions for complex scientific and technological challenges. As an Asian-Pacific American owned, for-profit organization structured as a partnership or limited liability partnership, Bayesics has established itself as a valuable partner in supporting NASA's research and development initiatives.

Through federal contract awards, Bayesics has primarily engaged with NASA's Shared Services Center to conduct research under the Small Business Innovation Research (SBIR) program. The company has been awarded several Phase I SBIR contracts, including a $149,853 project to develop a framework for Earth system digital twins with machine-learned parameterizations and a $149,983 project to create a spatially and temporally aligned POSIX-compliant data store for event-based analysis. These contracts, typically set aside for small businesses, demonstrate Bayesics' capabilities in developing cutting-edge technologies to support NASA's modeling and simulation needs. Additionally, Bayesics received a $253,710 contract to research a new snow observing strategy in support of hydrological science and applications. The company's most recent contract award, valued at $544,972, focuses on characterizing neural network classification uncertainty through Bayesian statistics, highlighting Bayesics' continued involvement in advanced data analytics projects for NASA.

Bayesics has also secured federal grant awards to conduct research and development projects for NASA. A notable example is a $209,529 Project Grant awarded in April 2024 under the Science (CFDA 43.001) program for characterizing neural network classification uncertainty through Bayesian statistics. This three-year project aims to develop advanced data analytics and modeling capabilities to support NASA's needs in areas such as Earth systems science and digital twins. Another grant of $253,710 was awarded in August 2022 for research on a new snow observing strategy, with a completion date of August 2024. As a subgrantee, Bayesics has contributed to a project led by DePaul University, where it provided expertise in developing the SpatioTemporal Adaptive-Resolution Encoding (STARE) system for precipitation feature extraction. In this role, Bayesics' team, led by Dr. Kwo-Sen Kuo and Dr. Michael L. Rilee, focused on STARE development and science analysis code development, serving as a crucial interface between software developers and scientists.