Research Scientist Intern

Pluralis Research works on Protocol Learning: training and serving large models in a fully decentralized way on small consumer-grade devices connected via the internet. Despite being dismissed as infeasible, we have made significant advances on this problem, most recently Agora, a permissionless run that pretrained an 8B model from scratch on consumer GPUs spread over the internet, with no single participant ever holding the full weights (tech report https://arxiv.org/abs/2607.13332). While many of the core research problems have been solved, Protocol Learning unlocks a series of new challenges. For the mission in full, read A Third Path: Protocol Learning https://pluralis.ai/blog/a-third-path-protocol-learning/.

This setting breaks nearly every assumption of datacenter training and inference: communication-efficient training across different parallelism axes, fault tolerance as nodes join and drop mid-run, heterogeneous compute and networks, and robustness to malicious participants. Our published methods include Subspace Networks https://arxiv.org/abs/2506.01260, Factored Gossip DiLoCo https://arxiv.org/abs/2606.22768, AsyncMesh https://arxiv.org/abs/2601.22442, and Sentinel https://arxiv.org/abs/2603.03592.

As a Research Scientist Intern you join us for a 6-month, fixed-term research internship during your PhD, focused on publishing. You work on the problems that stay open as we push from the 8B run toward frontier scale, with access to significant compute and focused mentorship from senior scientists. These are foundational papers up for grabs.

KEY RESPONSIBILITIES

• Publish in Tier-1 venues: Conduct novel research in Protocol Learning with the explicit goal of publishing in tier-1 ML conferences (NeurIPS, ICML, ICLR).

• Own a real problem: Pick a question that blocks Protocol Learning at scale and answer it — the internship is scoped so a foundational paper is a realistic outcome, not a stretch goal.

WHAT WE'RE LOOKING FOR

• Publication track...