Member Of Technical Staff

About

Sampura Research is an AI safety research non-profit focused on scalable oversight for frontier models through improved human+AI “judges”.

Judges (LLM, human, or a hybrid) are essential for aligning highly capable models through their use in evaluation and training. Limitations in today's judges can lead to undesirable behavior such as reward hacking in training, or failure to spot misalignment in evaluation. As models become more capable and their data more complex, often beyond what the systems overseeing them can understand, we expect these limitations to grow in severity and to become harder to notice.

We are designing novel judge protocols which leverage the strengths of both humans and models to outperform either alone, and building high-quality evaluations to explore judge performance across a diverse set of safety-related tasks. We believe that our focused approach will rapidly advance the science of judges by providing a much-needed systematic measurement of which protocols actually work, and producing state-of-the-art judges which can be used in real world training and evaluation pipelines.

Backed by an $11M grant from Coefficient Giving ($7M for the first year, and another $4M pledged), Sampura Research was founded in 2026 by members of DeepMind’s alignment and human data teams.

Please see our announcement for more information.

Role Overview

As a founding Member of Technical Staff, you will help shape the research and engineering direction alongside the founding team, and own end-to-end research and engineering problems within our agenda:

Evaluation: Build diverse and high-quality evaluations for judge protocols, spanning static datasets (existing and novel) and dynamic settings where policies actively optimize against the judge during RL

Methods: Research and implement strategies for improving human+AI judge protocols, such as learning-to-defer, AI-generated assistance for human raters, and co-training LLM judges alongside the policy during RL.

Infrastructure: Build the tools required to enable and scale our research agenda, including large-scale evaluation and training pipelines, a human rating platform, and a public judge leaderboard.

We've initially budgeted ~$1M per quarter for compute and human data, across a team of fewer than ten, to ensure that we can commit to ambitious goals in all of these areas. You'll scope ambiguous problems independently, communicate progress clearly, and synthesize findings into papers, blog posts, and benchmark releases.