Software Engineer focused on distributed systems, high-concurrency architectures, and AI-driven data pipelines. Experienced in building resilient, production-grade infrastructure at scale.
Designed a large-scale trend detection system processing ~20M records weekly. Pipeline includes embedding generation, clustering, summarization, and validation layers, enabling real-time insight extraction.
Developed a high-performance image classification API optimized for concurrency. Leveraged asynchronous FastAPI and Triton batch inference, reducing infrastructure costs by ~75% while improving response latency.
Engineered a vector-based deduplication engine using Qdrant, enabling high-speed similarity search and scalable content filtering with low-latency performance.
Built embedded firmware for health devices on nRF52832 using FreeRTOS. Implemented power-efficient scheduling and custom BLE services for real-time health data synchronization.