CS & Physics @ University of Chicago · MTS Intern @ Tensormesh
I build AI systems at the boundary of models, infrastructure, and evaluation, with a growing focus on autonomy and real-time inference.
Currently working on GPU management, LLM observability, inference infrastructure at Tensormesh.
Check out Hostess - a better way to deploy stacks!
- Tensormesh — integrated Phoenix into the observability stack with OpenInference tracing for LLM router traffic, plus SDK/CLI trace inspection and production rollout support.
- Hostess — Docker Compose for production.
- Redis Operator — Redis Kubernetes operator for managing instances at the pod level.
- AutoMoE — MoE-based self-driving system using PyTorch, CUDA, CARLA, and Hugging Face datasets.
- Launchpad — AI research matching platform with semantic search and LLM-based fit scoring.
