Job Details

Job Description

Pluralis Research works on Protocol Learning- training large models in a fully decentralised 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 trainingan 8B llama model from scratch with devices in physically different regions. While many of the core research problems have been solved, Protocol Learning unlocks a series of new challenges. If you want your work to shape the future of truly open innovation in the large model regime, join us.

Key Responsibilities

• Publish in Tier-1 Venues: At the core of Protocol Learning are hard research problems. There are foundational papers up for grabs. Solve these problems, and publish.

What We're Looking For

• Research Excellence: PhD in Machine Learning with publications in top-tier conferences (NeurIPS, ICML, ICLR).
• Distributed ML Experience: Exposure to distributed training, or federated learning.
• Implementation Skills: Strong programming abilities in PyTorch, with experience training models across multiple devices.
• Bonus: Experience with large language models, post training, RL etc.

Compensation & Benefits

• Equity-Heavy Package: We offer significant ownership or token allocations for key technical contributors in addition to high base salary.
• Visa Sponsorship: Optional full visa sponsorship and relocation support to either US or Australia.
• Remote-First Culture: Flexible work environment with team members distributed globally. We have two hubs; San Fransisc, and Melbourne, with the option for hybrid work if desired.
• Cutting-Edge Domain: Work at the intersection of AI and distributed systems, tackling some of the most challenging research problems in what is about to be one of the largest intersections of two previously non-overlapping fields ever.

FYI's

• We only hire in Australia and the United States. Visa sponsorship is limited to these countries.
• Applicants must have professional-level English proficiency (written and spoken).
• Pluralis is a remote team across Australia and the US. You'll need to be comfortable working across timezones and collaborating with a diverse, distributed group.
• Recruiters: we aren't looking for agency support at this time. We'll reach out if we need help.

We are backed by Union Square Ventures and are a world-class, deeply technical team of ML researchers. We view the world as a better place if we are able to implement what we are attempting, and Protocol Learning as the way to prevent corporations monopolising model development, access and release, and achieving massive economic capture. If this resonates, please apply.