Sean
Kraemer.
I’m a data scientist at Zurich North America in Chicago. I build AI tools that help underwriters review documents and trace answers back to their sources. Before that, I worked on insurance pricing models. Alongside work, I returned to the University of Illinois Urbana-Champaign and completed my Master of Computer Science in May 2026.
Featured work
All projects- 01
Distributed Stream Processor
A stream-processing system built from scratch in Go: SWIM failure detection, a replicated file system with consistent hashing, and exactly-once streaming on a 10-node Docker cluster.
Go · Docker
- 02
Devil's Advocate
A real-time AI debate platform that stress-tests startup ideas: an adversarial voice debater with retrieval memory, live judge scoring over Socket.IO, and a public demo.
React · FastAPI · Socket.IO · Gemini
- 03
AI Agent
A coding-agent harness with structured tools, session memory, and an evaluation runner. The entire loop is testable offline against a deterministic model.
Python · LLM tool use
- 04
Code-LLM Prompting Study
We tested prompting strategies on a small code model. In our code-generation experiments, cleaning up the output improved results more than changing the prompt.
Python · PyTorch · transformers
Latest writing
Building a distributed system from scratch
What I learned coding a stream processor from scratch in Go for a graduate distributed systems course, and how scripted live demos forced the system to be explainable, concise, and consistent.