Remote RL Post-Training Scientist — Decentralized AI Protocols

Pluralis-Research; San Francisco (CA); Remote; USD 180,000 - 260,000; Full time

Benefits offered by this job

Equity-heavy package

Remote-first culture

Visa sponsorship

Job summary

Pluralis Research is pursuing Protocol Learning, training and serving large models in a decentralized manner across consumer devices. The role focuses on building the RL post-training stack end-to-end, adapting algorithms for asynchronous, high-latency environments, and shipping the first decentralized post-trained models.

You will work in a remote-first global team, contributing to cutting-edge research with an equity-heavy base salary and visa sponsorship options.

Qualifications

• Hands-on RL post-training experience on large language models (RLHF/RLVR or related).
• Strong Python & PyTorch engineering skills with production-quality code.
• Experience with asynchronous, distributed RL and weight synchronization.

Responsibilities

• Build the post-training stack end-to-end: rollout ingestion, reward calc, policy updates, and weight propagation.
• Invent and adapt RL algorithms for asynchronous, high-latency, partially trusted generation.
• Ship the first decentralized post-trained models and validate them end-to-end.

Skills

RL post-training

Python

PyTorch

Distributed systems

Asynchronous training

Weight synchronization

Experimentation

vLLM

SGLang

Tools

Python

PyTorch

vLLM

SGLang

Distributed systems

Job description

Pluralis Research is pursuing Protocol Learning, training and serving large models in a decentralized manner across consumer devices. The role focuses on building the RL post-training stack end-to-end, adapting algorithms for asynchronous, high-latency environments, and shipping the first decentralized post-trained models.

You will work in a remote-first global team, contributing to cutting-edge research with an equity-heavy base salary and visa sponsorship options.