Pluralis Research
Research Scientist - Geo-Distributed Inference
San Francisco +1 · Remote · Full-time · Posted Oct 14, 2026
- Location
- San Francisco
Australia - Experience
- No specific number of years stated.
- Visa sponsorship
- Optional full visa sponsorship and relocation support to Australia or the US.
About the job
Build and operate a fast, reliable inference pipeline for reinforcement-learning training and future model serving across decentralized consumer hardware and public-internet connections.
What you'll do
- Own the inference stack end-to-end, including pipeline-parallel execution, placement and routing, transport, serving engine, and failure handling.
- Develop, validate, and deploy algorithms for fast inference on consumer hardware over the public internet.
- Keep the rollout pipeline reliable for RL training and evolve it into a serving layer for trained models.
What they ask for
- Have shipped serving-engine internals or built a large-scale inference system, and be able to do hands-on implementation.
- Professional-level English proficiency, written and spoken.
- Comfort working across time zones.
Nice to have
- Research publications or demonstrable unpublished work in distributed inference or a related field.
- Experience with low-bandwidth, high-latency systems; RL post-training; Apple silicon or MLX; P2P networking or NAT traversal; or AI labs.
Tools
Macs, consumer GPUs, MLX
AI focus
Develop decentralized inference systems that support RL rollouts and serve trained models across permissionless, trustless networks.
Model-written summary · the original posting is authoritative