Writing
Notes on AI systems, measurement, and reliability under uncertainty: agents, evaluation, anomaly detection, and adversarial ML.
2026
Mar 18, 2026
Extended Isolation Forest for Distributed Spark/Scala Anomaly Detection
Extended Isolation Forest support for LinkedIn's open-source Spark/Scala isolation-forest library, including random hyperplane splits, benchmarks, synthetic plots, and validation evidence. The work also became a useful case study in how to validate AI-assisted production code with evidence rather than trust.
Jan 9, 2026
Reinforcement Learning for Orbital Transfers at the 2026 AI Winter School (Brown University)
Hands-on workshop using reinforcement learning (PPO) to solve a simplified orbital transfer problem. Environment design, reward shaping, and training RL agents for space mechanics.
2025
Feb 10, 2025
Exploring LLMs and RAG at the 2025 AI Winter School (Brown University)
Hands-on workshop on LLMs and Retrieval-Augmented Generation for physics workflows, using the OpenAI API, an open Llama model, and LlamaIndex over research documents.
2024
Sep 23, 2024
Announcing ONNX Support in Isolation Forest
LinkedIn's open-source isolation forest library now supports ONNX export, enabling deployment beyond Spark for streaming and edge inference applications.
Aug 15, 2024
Finding AI-Generated Faces in the Wild
Detecting GAN- and diffusion-generated faces in the wild: one classifier across ten synthesis engines, robust to low resolution and heavy compression.
2023
Jun 20, 2023
Detecting AI-Generated Profile Photos
CVPR Workshop research on detecting GAN-generated profile photos from compact embeddings: a lightweight approach to catching synthetic faces at scale.
2021
Sep 2, 2021
Using deep learning to detect abusive sequences of member activity
Production deep learning approach for detecting abusive sequences of member activity at LinkedIn using request-level behavior streams.
2019
Aug 13, 2019
Open Source: Spark/Scala Isolation Forest Library
Announcement and resources for LinkedIn's open-source Scala/Spark isolation-forest library for large-scale unsupervised anomaly detection.
2016
Oct 7, 2016
Detection of unusual travel patterns to prevent user account compromise
Insight Data Science consulting project using unsupervised anomaly detection to flag unusual travel patterns and possible account compromise.
Aug 18, 2016
Calibrating the LUX Dark Matter Experiment
Ph.D. thesis work: an absolute, in situ calibration of sub-keV nuclear recoils in the LUX dark matter detector using D-D neutron scattering kinematics.