Summary
Raghavendra Vedula is a machine learning and systems engineer with nine years of experience building production-grade inference and data pipelines for speech, vision, and finance automation. He’s worked across startups and hyperscalers—optimizing LLM/GPU inference at Amazon for 10x throughput gains, developing long-context generation stacks, and now focusing on LLM/VLM inference at Fireworks AI. As a founding engineer he built document-understanding models for financial automation, blending spatial attention and BERT-style architectures with scalable training and serving pipelines. His background includes low-power wakeword models, large-scale speech processing with Apache Beam, and on-device AR metrics work at Google, showing both edge and cloud expertise. Based in California with an MS in Computer Science from USC, he combines deep performance optimization skills with a talent for turning research ideas into robust, deployable systems.
9 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 (Honors), Computer Science and Engineering, Bachelor of Technology - BTech (Honors), Computer Science and Engineering at Odisha University of Technology and Research
Kendriya Vidyalaya Sangathan
English, Hindi, Telugu