Remote Distributed Inference Systems Architect

Pluralis Research; California; Remote; USD 180,000 - 240,000; Full time

Benefits offered by this job

Equity-heavy package

Remote-first culture

Visa sponsorship and relocation

Job summary

Pluralis Research is advancing Protocol Learning by training and serving large models in a decentralized way on consumer devices connected via the internet. Our work enables RL rollouts today and serving trained models in a permissionless, trustless setting.

We are building the end-to-end inference stack, including pipeline-parallel execution, placement and routing, transport, and a robust serving engine designed to operate under variable network conditions and weight updates.

Qualifications

• Shipped serving-engine internals or large-scale inference systems experience.
• Experience with networks and systems in low-bandwidth, high-latency environments is a plus.
• Research publications or unpublished work in distributed inference or LLM serving systems beneficial.

Responsibilities

• Own the inference stack end-to-end: pipeline-parallel execution, placement, routing, transport, serving engine, and failure handling.
• Invent the algorithms to make inference faster on consumer hardware over the public internet and validate them in production.
• Serve training rollouts and users: keep the rollout pipeline fast and reliable for RL training and future serving of trained models.

Skills

Shipped serving systems

Low-bandwidth networking

Distributed systems

RL familiarity

Job description

Pluralis Research is advancing Protocol Learning by training and serving large models in a decentralized way on consumer devices connected via the internet. Our work enables RL rollouts today and serving trained models in a permissionless, trustless setting.

We are building the end-to-end inference stack, including pipeline-parallel execution, placement and routing, transport, and a robust serving engine designed to operate under variable network conditions and weight updates.