Diwakar Singh@diwakar_v_singh820
0
Events attended
0
Submissions made
United States
5+ years of experience
About me
I’m a Machine Learning Engineer focused on reasoning models, not just making them smarter, but making them trainable, measurable, and shippable. Currently, I work on the Amazon Nova foundation model stack and the systems that power post-training and reasoning optimization at scale: RL/post-training pipelines, standardized training environments (packages + images), and high-throughput evaluation for step-wise reasoning quality. My niche is the glue between scientists and systems: I translate research requirements into reliable pipelines, and I translate system constraints into workflows researchers can iterate on quickly. Previously, I worked on search and Document AI at scale, and I’m drawn to roles in LLM/RL training, reasoning optimization, evaluation/benchmarking, and large-scale ML systems.
