Decentralized RL Post-Training Scientist

OpenKrill; United States; Remote; USD 180,000 - 280,000; Full time

Benefits offered by this job

Equity-heavy package

Remote-first culture

Visa sponsorship and relocation to US

Relocation support to Australia

Job summary

Pluralis Research is hiring a specialist to advance Protocol Learning by building RL post-training stacks that run across consumer GPUs and the public internet. You will adapt RL algorithms to asynchronous, high-latency, decentralized settings, and ship first post-trained models.

The role demands strong Python/PyTorch engineering, distributed systems expertise, and a passion for protocol learning. You will work with a global team to push the boundaries of decentralized AI, delivering robust

Qualifications

• RL post-training on large language models and systems-level work.
• Experience with rollout generation and async training loops.
• Publications in RL/post-training or related fields are a plus.
• Ability to work across time zones and with remote teams.

Responsibilities

• Build end-to-end RL post-training stack including rollout ingestion and policy updates.
• Develop asynchronous, high-latency training pipelines and weight synchronization.
• Evaluate and release first decentralized post-trained models.

Skills

Python

PyTorch

Distributed systems

RL post-training

English proficiency

Tools

vLLM

SGLang

Git

Job description

Pluralis Research is hiring a specialist to advance Protocol Learning by building RL post-training stacks that run across consumer GPUs and the public internet. You will adapt RL algorithms to asynchronous, high-latency, decentralized settings, and ship first post-trained models.

The role demands strong Python/PyTorch engineering, distributed systems expertise, and a passion for protocol learning. You will work with a global team to push the boundaries of decentralized AI, delivering robust