James Verbus

AI systems for
adversarial settings.

At LinkedIn, I worked on automation detection and AI-generated profile photos, and built the Spark/Scala isolation-forest library.

Read posts Representative work

James Verbus working on the LUX dark matter detector — LUX dark matter detector, Sanford Underground Research Facility.

What I work on

Representative work

Sequence modeling

Deep learning for abusive activity sequences

Detecting low-volume scrapers from the order and timing of their requests.

Research

Finding AI-generated faces in the wild

Testing a face detector on GAN and diffusion models, including generators withheld from training.

Workshop

Reinforcement learning for orbital transfers

Training PPO policies for orbital transfers and comparing them with the Hohmann solution.

Open source anomaly detection

Open-source Spark/Scala isolation-forest library

How random partitions identify unusual account activity, and how to distribute training and scoring in Spark.

View all writing

Career

Background

At LinkedIn, most recently as a Senior Staff Machine Learning Engineer, I built production detection systems for Trust. I created and open-sourced the Spark/Scala isolation-forest library and led AI productivity work for Trust engineering teams.

Before LinkedIn, I earned my Ph.D. in physics at Brown working on LUX, one of the world's most sensitive dark-matter detectors. The common thread is practical measurement: extracting weak signals from noisy data, checking uncertainty, and making systems useful when ground truth is incomplete.