James Verbus
AI and physics
I was a Senior Staff Machine Learning Engineer at LinkedIn, where I built AI systems to detect bots and abusive automated activity. I also developed the open-source Spark/Scala isolation-forest library.
Before that, I earned my Ph.D. in physics at Brown, working on calibration of the LUX dark matter detector.
Selected work
-
Isolation forest for Spark/Scala
Distributed unsupervised machine learning for anomaly detection with Isolation Forest and Extended Isolation Forest.
-
Finding AI-generated faces in the wild
Evaluating deep learning models for detecting AI-generated faces across GAN and diffusion engines, including generators withheld from training.
CVPR Workshop on Media Forensics
-
Deep learning for abusive activity sequences
Using deep learning and learned request embeddings to detect low-volume scrapers from activity sequences and timing.
LinkedIn Engineering
-
Calibrating the LUX dark matter experiment
Measuring the charge and light response of liquid xenon to low-energy nuclear recoils using D-D neutron scattering in LUX.
Brown University · LUX