Summary
Suhas Maringanti is a Machine Learning Engineer and Applied Scientist with over a decade of experience building and optimizing AI systems across industry and research. He has 3+ years focused on LLM inference, agentic RL training recipes, post-training data curation, and model evaluation, with hands-on work deploying inference optimizations like speculative decoding and parameter-efficient finetuning at AWS. At Amazon he helped build Rufus, an AI shopping assistant, contributing to multimodal data curation, safety, and evaluation, and now leads platform work for efficient LLM inference at d-Matrix. His background spans systems-level research—accelerating combinatorial algorithms on supercomputers—to practical product engineering, giving him a rare blend of deep research rigor and production-first pragmatism. Suhas holds a PhD in Engineering and Applied Science and complements his research roots with experience across the full ML lifecycle, from NAS for latency-sensitive models to real-world RAG and prompt routing. Colleagues rely on him for bridging prototype research into scalable, safety-conscious AI deployments.
11 years of coding experience
4 years of employment as a software developer
MPC(Mathematics, Physics and Chemistry) with English, MPC(Mathematics, Physics and Chemistry) with English at Narayana Junior College
Doctor of Philosophy - PhD, Engineering and Applied Science, Doctor of Philosophy - PhD, Engineering and Applied Science at University of Massachusetts Dartmouth
Bachelor of Science (BS), Computer Science, Bachelor of Science (BS), Computer Science at Missouri State University
English, Telugu, Hindi