Machine Learning Engineer - ML Training Platform

Pluralis Research
San Francisco
Workplace: RemoteFull timeFunction: Data Science & Machine LearningExperience: 5+ yearsSkills: ["Python","Docker","Kubernetes","EKS","Pulumi","Terraform","CloudFormation","Prometheus","Grafana","NVIDIA runtime"]

Architect and build the core infrastructure for a decentralized ML training platform, enabling multi-cloud deployment, distributed training, and real-world network condition testing. You’ll own infrastructure orchestration, GPU workloads, and data management across AWS, GCP, and Azure, driving fault-tolerant, scalable systems that support continuous experimentation and large-scale model training.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Pluralis Research
Pluralis Research
4 months ago

Machine Learning Engineer - ML Training Platform

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 11 hours agoStatus: Live

Job Summary

Architect and build the core infrastructure for a decentralized ML training platform, enabling multi-cloud deployment, distributed training, and real-world network condition testing. You’ll own infrastructure orchestration, GPU workloads, and data management across AWS, GCP, and Azure, driving fault-tolerant, scalable systems that support continuous experimentation and large-scale model training.
Location: San Francisco
Workplace: Remote
Employment Type: Full time
Job Function: Data Science & Machine Learning

Key Responsibilities

  • •Multi-Cloud Infrastructure: 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.
  • •Distributed Training Systems: Architect fault-tolerant infrastructure for distributed ML, including GPU clusters, NVIDIA runtime, S3 checkpointing, large dataset management and streaming, health monitoring, and resilient retry strategies.
  • •Real-World Networking: Build systems that simulate and handle real-world network conditions—bandwidth shaping, latency injection, packet loss—while managing dynamic node churn and ensuring efficient data flow across workers with heterogeneous connectivity.
  • •Scale & Reliability: Ensure self-healing, observability, and incident response in production-grade ML training platforms, coordinating across multiple services and teams.
  • •Collaboration & Technical Leadership: Partner with ML researchers and infrastructure engineers to iterate rapidly on platform features and reliability

Key Requirements

  • •5+ years of experience in infrastructure & platform engineering with IaC (Pulumi/Terraform/CloudFormation)
  • •Production multi-cloud deployments with Docker/Kubernetes (EKS), GPU workloads, and heterogeneous clusters at scale
  • •Distributed training workflows: checkpointing, data sharding, model versioning, long-running job orchestration, and decentralized networking (P2P, NAT traversal)
  • •Strong Python engineering (asyncio, concurrency, retry logic, cloud SDKs, CLI tooling) with observability, SRE practices, monitoring (Prometheus/Grafana)
  • •Experience in a startup environment with emphasis on micro-services orchestration or big tech background
Experience:5+ yearsMulti-cloudDistributed systemsMl infrastructure
Skills:PythonDockerKubernetesEKSPulumiTerraformCloudFormationPrometheusGrafanaNVIDIA runtime
Languages:English
Tech Stack:PythonDjangoFlaskAWSGCPAzureKubernetesDockerPulumiTerraformCloudFormationPrometheusGrafanaNVIDIAS3ETLSDKs

Company Brief

Pluralis Research
Develops Protocol Learning for decentralized, multi‑participant training of foundation models so models remain unmaterialized and community‑owned, enabling open-source large‑scale AI without single‑party control.
Industry: AI & Machine Learning
Company Size: Micro (1 to 10 employees)
Growth: Early Stage Startup
Funding: Seed
Founded: 2024
Glassdoor
Glassdoor: 4.0
WebsiteLinkedIn