Software Engineer and Machine Learning Researcher. Building efficient AI systems, developer tools, and performance-oriented software.
Selected work
A few projects that show how I approach efficiency, reliability, and system design.
Low-rank and activation-aware compression for language and vision-language models, evaluated for quality, memory, and speed.
Agentic security-analysis workflows backed by an MCP server, retrieval, and specialized tools for network and system telemetry.
A C++ matching engine maintaining bid/ask books and processing add, cancel, modify, and execute events with price-time priority.
Experience
TODO – TODO
TODO
Built agentic security-analysis tooling on top of an MCP server, connecting retrieval and specialized tools for network and system investigation workflows.
TODO – Present
New York, NY
Researching efficient inference for language and vision-language models, focused on compression methods that preserve task quality.
TODO – TODO
TODO
Built web platform features for online classrooms and educational workflows.
Current focus
Efficient inference for language and vision-language models
Performance-aware systems programming
Reliable agent and developer tooling
Technical notes
No published notes yet
Notes in progress will appear here once finished.