Summary
Rahul Sharma is a machine learning engineer with nearly a decade of experience building and productionizing deep learning and RL systems across major tech firms including Meta and Reddit. He specializes in speech, vision, NLP and GPU-accelerated training, and has a track record of translating research into pragmatic frameworks that cut experimentation time and lower deployment cost. At Jio he developed phonetic embeddings that beat SOTA by 12% and replaced MFCC features to improve ASR in noisy conditions, while at Liv.ai he engineered cross-platform C++ tooling to port TF models to mobile and server targets. Equally comfortable in research and engineering roles, he has published academic work, holds IBM certifications, and blends systems-level C++/performance optimization with applied ML for high-impact consumer and ads products. Based in the UK, he combines an RL enthusiast’s curiosity with a system developer’s discipline to move models from prototype to millions-of-users scale.
9 years of coding experience
10 years of employment as a software developer
Master of Engineering (M.E.) Computer Science, Master of Engineering (M.E.) Computer Science at Indian Institute of Science (IISc)
Bachelor of Technology (B.Tech.) Computer Science, Bachelor of Technology (B.Tech.) Computer Science at COER University
English, Hindi