After being diagnosed with a meningioma and p1RCC (a rare kidney cancer) in 2014, Bill co-founded rarekidneycancer.org with Dr. James Hsieh in 2016. In 2018 and 2020, with Pete Kane, he brought together 17 teams of more than 100 researchers who used his medical data to determine “genes of interest” related to his disease. He applied ensemble reasoning to evaluate the results of these hackathons (which Bill calls “gamified tumor boards” ) to discover better ways to do cancer research. Bill is currently automating these “gamified tumor boards” using Large Language Models (LLMs). Bill is also involved in several patient organizations and is a CDMRP reviewer.
Prior to rarekidneycancer.org, Bill worked with Sabrina Paseman to use non-invasive Blood fluorescence to detect iron deficiency anemia (Patent 8,306,594) and with Katherine Paseman to expand Pulse oximetry technology to do non-invasive measurement of Hemoglobin and Hematocrit.
Bill’s 50 year career in AI started at age 18, when he worked with Howie Johnson at Rice University using n-grams (like chatGPT) to do classical music composition. He continued his music work at MIT in Steve Ward’s lab exploring generative grammars and constraints. He then applied his Symbolic AI knowledge to Electrical Design Automation as the 16th employee at Daisy Systems. Bill next founded Atherton Technology (which failed) and then Calico Commerce (which IPOed), where he repurposed his music composition system to do sales configuration over the internet (Patent 5,745,765). After Bill took Calico public, he retired and, as discussed above, has dedicated his time to rarekidneycancer.org, patient advocacy and working with his daughters on non-invasive blood analysis.