Evan Ochsner
Scientist. Builder. Physics PhD; eight years as a lead applied scientist at Amazon. I work on hard problems with messy data and ship something useful at the end.
About
Physics PhD (Maryland), trained at Chicago. Started in gravitational-wave data analysis for LIGO — post-Newtonian theory, matched-filter search pipelines, co-leading one of the flagship searches.
Pivoted to cybersecurity at Sqrrl, a Boston startup spun out of NSA threat-hunting work, where I built detection and analytics on real customer data at enterprise scale. Sqrrl was acquired by Amazon, and I stayed on for eight years as an applied scientist — one of four scientists who launched Amazon Detective, then working in the Security Analytics and AI Research (SAAR) team as lead scientist on major Detective feature additions and the relaunch of AWS Security Hub.
I like problems where the data is messy, the answer matters, and the work has to actually run in production.
Currently searching for my next problem to tackle.
If you have one, reach out on LinkedIn.
Projects
Transformers: A Guided Tour
Transformers are likely to be the defining technology of our time. My attempt to understand how they work, explain it to a lay audience, and frame it in a way that's still interesting to experts. Easy. Heavy on interactive visualizations and metaphor — an intuitive model for what transformers are actually doing.
ToolkitPersonal Advocacy Toolkit
For people getting stonewalled by an insurer, a provider, a landlord, a debt collector. Evidence intake with forensic integrity, drafting tools with anti-hallucination guard rails, a publication-safety pipeline — does the legwork so whoever helps you can actually help you.
DataU.S. Insurance Complaint Data
Normalized complaint and litigation rate data from 20+ state insurance regulators, with an interactive multi-state viewer for cross-state comparisons and trend analysis.