Hello! I'm a software engineer who has previously built bond trading exchanges, financial reconciliation systems, and EV charging networks.
Currently exploring efficient multi-modal ML inference systems.
LLM serving engine with Go control layer, gRPC Python workers, priority scheduling, rate limiting, dynamic and continous batching.
Demand-response protocol for GPUs that curtails power caps in real time via WebSockets and nvidia-smi
System that tracks crypto AI agents token recommendations on Twitter and ranks them by prediction accuracy