Research Engineer - Post-Training
AI
Pluralis Research is seeking a Research Engineer for Post-Training to build and invent RL post-training algorithms and systems for decentralized training on consumer-grade devices. The role involves end-to-end ownership of the post-training stack, from rollout ingestion to weight synchronization, and requires hands-on experience with RL post-training on large language models, strong Python and PyTorch engineering skills, and a research mindset. This is a full-time, on-site position based in Australia with visa sponsorship and relocation support available.
What you'll do
- Build the post-training stack: You build the RL training loop end-to-end: rollout ingestion from the geo-distributed inference pipeline, reward computation, policy updates, and getting updated weights back out to the network. You set the direction, and you make things happen.
- Invent the algorithms: Standard RL recipes assume on-policy rollouts from fast, trusted hardware. You adapt them to asynchronous, high-latency, partially trusted generation: staleness tolerance, off-policy corrections, and communication-efficient policy updates.
- Ship first post-trained models: You build the evals that show the models are improving, and you take the first decentralized post-trained release from run to public artifact.
What you need
Must-have
PythonPyTorchRL post-trainingRLHFRLVRReasoning-focused RLRollout generationAsync training loops
Nice-to-have
Slow networks trainingDecentralized setupsFederated setupsvLLMSGLangReward modelingVerifiable-reward datasetsP2P networkingNAT traversal
Soft skills
Mission alignmentResearch abilityStrong engineeringComfortable working across timezones
Research EngineerPost-TrainingRLDecentralized AIProtocol LearningPythonPyTorchFull-timeOn-siteAustralia