Rituraj

Rituraj

I work on verifier-guided inference, reasoning evaluation, and test-time search for language models.

Currently: building PRISM and measurement systems for long-horizon reasoning.

Read PRISMPRM-guided inference for Deep Think · arXiv

X logo GitHub logo YouTube logo LinkedIn logo

/recent

Apr 2026ghostenv: Stop Leaking Secrets to AI AgentsYour .env file is the first thing an AI agent reads. It has to. That’s where the config lives.[security, devtools, ai, cli, open-source]Mar 2026PRISM in 3 Parts: Scoring, Resampling, and Accept/Reject FilteringMost parallel test-time scaling pipelines follow the same pattern: generate a pool of candidate…[reasoning, inference, prm, search, systems]Mar 2026PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided InferenceTL;DR PRISM uses step-level verification (a PRM signal) to guide population refinement and solu…[research, reasoning, inference, PRM, deep-think]

/index

~/research2 itemsPapers and preprints~/projects1 itemPersonal and open source projects~/blog3 itemsBlog posts and articles

Press cmd+k to jump anywhere.

Last updated Apr 22, 2026