Pluralis Research

• Write blog posts, threads, and explainers for ML engineers and researchers. • Engage with the community on Discord and Twitter/X. • Represent Pluralis at ML conferences by giving talks and running demos. • Improve onboarding and communication for GPU contributors. • Produce and edit multimedia content for YouTube. • Create co-marketing materials with ecosystem partners. • Track metrics across channels and optimize engagement. • Provide market intelligence on decentralized training and model development.

• Architect, build, and scale the foundational infrastructure powering our decentralized ML training platform • Design resource management systems provisioning and orchestrating compute across AWS, GCP, and Azure using infrastructure-as-code (Pulumi/Terraform) • Handle dynamic scaling, state synchronization, and concurrent operations across hundreds of heterogeneous nodes • Architect fault-tolerant infrastructure for distributed ML including GPU clusters, health monitoring, and resilient retry strategies • Build systems that simulate and handle real-world network conditions

• Design and implement large-scale distributed training systems optimized for heterogeneous hardware operating under low-bandwidth, high-latency conditions. • Develop and optimize model-parallel training strategies (data, tensor, pipeline parallelism) with custom sharding techniques that minimize communication overhead. • Optimize GPU utilization, memory efficiency, and compute performance across distributed nodes. • Implement robust checkpointing, state synchronization, and recovery mechanisms for long-running, fault-prone training jobs. • Build monitoring and metrics systems to track training progress, model quality, and system bottlenecks. • Architect resilient training systems where nodes can fail, networks can partition, and participants can dynamically join or leave. • Design and optimize peer-to-peer topologies for decentralized coordination across non-co-located nodes. • Implement NAT traversal, peer discovery, dynamic routing, and connection lifecycle management. • Profile and optimize communication patterns to reduce latency and bandwidth overhead in multi-participant environments.

• Technical Content: Write blog posts, threads, and explainers that break down our research for ML engineers and systems researchers. • Community Engagement: Own Discord and Twitter/X as technical channels. Engage directly with researchers and engineers in ML and decentralized compute communities. • Conference Presence: Represent Pluralis at ML conferences. Give talks and build relationships with the research community. • Contributor Experience: Improve onboarding for GPU contributors. Write documentation and guides. • Multimedia: Produce and edit video content—paper walkthroughs, research explainers, contributor tutorials. Manage the YouTube channel. • Partnership Support: Create co-marketing materials and coordinate announcements with ecosystem partners. • Growth and Analytics: Track metrics, run experiments, and own reporting on community growth and content performance. • Market Intelligence: Track developments across decentralized training and feed insights to leadership.

• Design resource management systems provisioning and orchestrating compute across AWS, GCP, and Azure. • Handle dynamic scaling, state synchronization, and concurrent operations across hundreds of heterogeneous nodes. • Architect fault-tolerant infrastructure for distributed ML. • Build systems that simulate and handle real-world network conditions.

• Manage Discord, build Twitter/X presence, and engage in relevant technical communities. • Turn research into blogs, threads, and videos, translating for different audiences. • Build email lists, create run a newsletter, optimize media channels. • Track metrics across channels, run experiments, and optimize for discoverability. • Create co-marketing materials and coordinate announcements with ecosystem partners. • Monitor ecosystem trends and feed positioning insights to leadership.

• Publish in Tier-1 Venues: At the core of Protocol Learning are hard research problems. There are foundational papers up for grabs. Solve these problems, and publish.

• Contribute to groundbreaking research in Protocol Learning during your PhD. • Join Pluralis for a 6-month research internship focused on publishing. • Conduct novel research within the domain of Protocol Learning, with the explicit goal of publishing in tier-1 ML conferences (NeurIPS, ICML, ICLR).