Distributed RL Scientist: Post-Training in Decentralized AI

pluralis-research; United States; Remote; USD 150,000 - 210,000; Full time

Benefits offered by this job

Equity-Heavy Package

Remote-First Culture

Visa Sponsorship & Relocation

Open Problems

Job summary

Pluralis Research is pioneering Protocol Learning, training and serving large models in a decentralized way across consumer hardware. We are hiring for hands-on RL post-training and delivery of first decentralized releases to the network.

In this role you will build end-to-end RL pipelines, optimize asynchronous training, and ensure weight synchronization across geo-distributed systems, while contributing to publications and system design for scalable serving.

Qualifications

• Hands-on RL post-training on large language models and systems work.
• Strong production-quality Python and PyTorch engineering.

Responsibilities

• Build the post-training RL training loop end-to-end: rollout ingestion, reward computation, policy updates, and weight propagation.
• Invent algorithms for asynchronous, high-latency, partially trusted generation with staleness tolerance and off-policy corrections.
• Ship first post-trained models and release public artifacts with evaluation pipelines.

Skills

RL post-training

Python programming

PyTorch

Distributed training

Async training

Weight synchronization

Tools

Python

PyTorch

Distributed training frameworks

RL libraries

Job description

Pluralis Research is pioneering Protocol Learning, training and serving large models in a decentralized way across consumer hardware. We are hiring for hands-on RL post-training and delivery of first decentralized releases to the network.

In this role you will build end-to-end RL pipelines, optimize asynchronous training, and ensure weight synchronization across geo-distributed systems, while contributing to publications and system design for scalable serving.