Research Engineer, Post-Training (Fully Remote)

Pluralis Research; United States; Remote; USD 170,000 - 230,000; Full time

Job summary

Pluralis Research is seeking an AI engineer to advance reinforcement learning systems. You will build the RL post-training stack end-to-end, from rollout ingestion to reward computation and policy updates, and you will implement robust monitoring and versioning to keep models auditable.

You will invent algorithms for asynchronous, high-latency, partially trusted generation, including staleness tolerance and off-policy corrections, and you will design experiments that validate improvements under

Responsibilities

• Build the RL post-training stack end-to-end, from rollout ingestion to reward computation and policy updates, with robust monitoring and versioning for auditable models.
• Invent algorithms for asynchronous, high-latency, partially trusted generation, including staleness tolerance and off-policy corrections, and design experiments under production constraints to validate improvements.
• Ship the first decentralized post-trained models, building evals and taking the release from run to public artifact, while collaborating with researchers and engineers for reproducible, scalable deployment.

Job description

• - Build the RL post-training stack end-to-end: rollout ingestion, reward computation, policy updates, and returning updated weights to the network.
• - Invent algorithms for asynchronous, high-latency, partially trusted generation, including staleness tolerance and off-policy corrections.
• - Ship the first decentralized post-trained models, building evals and taking the release from run to public artifact.