SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Pluralis Research is building Protocol Learning—a system for training and serving large language models in a fully decentralized way on consumer-grade GPUs connected via the internet. The company recently demonstrated this with Agora, a permissionless pretraining run that trained an 8B model from scratch on distributed consumer hardware with no single participant holding full weights.
In a trustless, permissionless network of distributed workers, verification is critical. Workers can cheat by returning tokens from cheaper models, using wrong sampling parameters, dropping updates, flooding the system, submitting poisoned gradients, or inflating their reported contributions. This role focuses on building efficient algorithms and systems to verify that training and inference work is legitimate.
Key responsibilities include: (1) owning and evolving the threat model across pretraining, post-training, and inference—covering training disruption, free-riding, model poisoning, data extraction, and reward manipulation; (2) designing and calibrating statistical verification methods with rigorous benchmarks and controlled error rates across heterogeneous hardware (GPUs, Macs); and (3) shipping and operating the verification service in the inference path, taking ownership of its real-world performance.
You'll need deep expertise in verification systems (shipped or published), statistical rigor in designing experiments and calibrating decision thresholds, and technical knowledge of the solution space for verifying untrusted compute—from statistical testing to re-execution, cryptographic proofs, and trusted hardware. Experience in fraud detection, anti-cheat systems, experimentation platforms, or published work on inference/training verification is valued equally with academic papers.
Nice-to-have skills include familiarity with large-scale pretraining and RL post-training, decentralized ML security and adversarial threat models (poisoning, Sybil, collusion, replay), and experience at proprietary, open-weight, or open-source AI labs.
The company is backed by Union Square Ventures and operates a remote-first culture with teams distributed across Australia and North America. Visa sponsorship and relocation support to either region are available. You'll be working on open research problems with few published answers, contributing to foundational work in decentralized AI.
