Summary
Reza Baharani is a Lead LLM Engineer and Deployment specialist with ~8 years of experience building power-efficient, real-time AI systems for edge devices and cloud-backed ML services. He combines a PhD in Computer Architecture with hands-on expertise in mobile and embedded deployment—optimizing 3D pose models to run at up to 20 FPS on device SoCs and mapping model shards across NPU/CPU/GPU. Reza architects scalable ML backends and microservices (gRPC, FastAPI, Ray) and runs production reasoning stacks using AWS tooling including Amazon Bedrock, Cognito, DynamoDB, and S3. His research background in self-supervised transformers, dVAE tokenization, and HW/SW co-acceleration informs practical optimizations like quantization, distillation, and scheduler-driven inference placement. Equally comfortable in Swift, React Native, TensorFlow Lite, CoreML and FPGA/C/C++ implementations, he bridges research and production to deliver constrained-system performance gains. A less obvious strength is his track record of squeezing high-end deep learning into tiny footprints—evidenced by multi-camera Full HD pipelines and Cortex-M7 deployments that maintain real-time throughput.
8 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Architecture, Doctor of Philosophy - PhD, Computer Architecture at University of North Carolina at Charlotte