Research Member of Technical Staff - Training Platform

Responsibilities

• Build and maintain training orchestration systems for large-scale distributed model training across GPU clusters
• Develop experiment management tooling including job configuration, tracking, reproducibility, and artifact management
• Build observability infrastructure for training runs such as loss curves, compute utilization, and anomaly detection
• Optimize and automate the research iteration loop from experiment launch to results analysis
• Manage job scheduling and cluster utilization for efficient use of GPU compute
• Build internal tooling and interfaces to accelerate researcher workflows
• Collaborate with training systems, data infrastructure, and research teams

Requirements

• Strong software engineering skills with experience in MLOps or ML platform engineering
• Familiarity with distributed training frameworks like PyTorch DDP, FSDP, DeepSpeed, or Megatron
• Experience building experiment tracking, reproducibility, and artifact management systems
• Experience managing and operating GPU cluster environments such as Slurm or Kubernetes
• Strong reliability engineering instincts including monitoring, alerting, and failure recovery

Preferred Qualifications

• Experience with training orchestration tools like Ray
• Familiarity with experiment tracking tools like Weights & Biases or MLflow
• Experience supporting large model training pipelines for LLMs, VLMs, or video models
• Understanding of parallelism strategies and training efficiency
• Experience with cloud-based training infrastructure (AWS, GCP, or Azure)

About the Company

Rhoda AI is building the next generation of generalist intelligent robots, owning the full robotics stack from high-performance hardware to state-of-the-art foundation world models. Having raised over $450M, the company is investing aggressively in model research, infrastructure, and manufacturing scale-up.