I lead backend development for Leo, Brave Browser's built-in AI assistant. I develop an AI chat server using Python and FastAPI, utilizing RAG to integrate real-time search and news results for contextually enriched, up-to-date responses. I also deploy and manage open-source LLMs, such as Llama 3 and Mixtral 8x7B, on Kubernetes using vLLM.

I delivered major features for the Web3 and Browser AI teams, contributing to both the Chromium-based browser in C++ (shipped to over 80 million users), and backend services in Golang and Python.

I developed core features for Brave Browser's built-in crypto wallet, including NFT support, hardware wallet integration, and user asset discovery using C++. I also led the development of backend APIs in Golang, providing crypto market data to the wallet

I led a team of 5 engineers in designing and implementing a rules-based classification framework in Python/SQL, preventing $100M in fraud over two years. I built data pipelines using Python and Kafka to consolidate tens of billions of records from 8 Postgres instances into a centralized Redshift data warehouse, enabling holistic analysis.