Summary
Muthu A is a seasoned software engineer with 11 years of experience, currently optimizing LLM inference at Amazon to power the shopping assistant Rufus with low-latency, high-throughput production traffic (~1,500 TPS). He has driven measurable impact—boosting system throughput by 30%—by implementing continuous batching, streaming responses, and building benchmarking/profiling tools that exposed key deployment optimizations. Prior work spans GNN-driven ranking and behavior-transfer techniques that contributed to multi-million dollar revenue gains, plus infrastructure migrations and abuse-reordering systems that improved platform trust. With an MS in Computer Science from USC and hands-on experience with vLLM, AWS Neuron, and ANN search, he blends research rigor (computational biology and remote sensing projects) with production-scale ML systems engineering. Muthu’s uncommon mix of academic research and large-scale deployment experience helps bridge model-level insights to robust, observable inference pipelines.
11 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at University of Southern California
Bachelor of Technology - BTech, Information Technology, Bachelor of Technology - BTech, Information Technology at College of Engineering, Guindy
English, Tamil, Hindi