Jeremy Nixon
Founder & CEO @ Infinity
San Francisco Bay Area · 2 connectionsInfinity Artificial Intelligence Institute
Not on Unlinked · Imported profile. Is this you? Claim it
Is this about you? Request removal
About
Dedicated to understanding the principles underlying information and intelligence.
Experience
Oct 2025
Automating AI Research. Accelerating inference.
Aug 2023
Sep 2022 to Aug 2023
Feb 2022 to Sep 2022
Jun 2025
Overseeing Omniscience IP, Licensing, & Corporate Decisions.
Apr 2021 to Jun 2025
Omni is a stunning immersive information retrieval system which puts you in immediate touch with all of your team & company's knowledge, as well as knowledge contained in books, on the web and in your personal notes and writings, no matter where you are in your browser. Omni generates text using your working environment and your company's existing data as context for writing emails, creative ideation, improved critical thinking, enhanced problem-solving and many more use cases.
Jan 2018 to Apr 2021
Google Scholar Articles: Title: Measuring Calibration in Deep Learning (2019) Title: Understanding and correcting pathologies in the training of learned optimizers (2019) Title: Analyzing the role of model uncertainty for electronic health records (2020) Title: Uncertainty baselines: Benchmarks for uncertainty & robustness in deep learning (2021) Title: What are you optimizing for? aligning recommender systems with human values (2021) Title: Why are bootstrapped deep ensembles not better? (2020) Title: Resolving spurious correlations in causal models of environments via interventions (2020) Title: Training optimizer neural networks (2020)
Jan 2016 to Nov 2017
· Implemented Deep Learning library on Spark, including Convolutional and Feedforward Neural Networks– a native distributed implementation in MLlib, integrated with present ANN structure. Implemented associated optimization and regularization tools. https://github.com/JeremyNixon/sparkdl · Original Author of Deep Neural Network Regression in MLlib. Implemented efficient distributed gradient computations. · Taught internal course on Gradient Boosting, Random Forests, Linear Models, and optimization. · MLConf, Dataworks, and Apache Big Data conference speaker on machine learning development, deep learning models and optimization.
Dec 2012 to Sep 2013
Designed, developed and launched an iPhone mobile application as a fellow at Harvard's Innovation Lab. TimeStamp uses type-free data entry to make it easy for users to track and aggregate their time usage.
Jun 2012 to Aug 2012
Built a computer cluster for molecular dynamics simulations; investigated electrical properties of single-walled carbon nanotubes.
Education
B.A. Applied Mathematics, Computer Science and Economics
Jan 2011 to Jan 2015
Summer Mathematics Institute, Mathematics
Jan 2009 to Jan 2011