Member of Technical Staff, Backend

About the Company

NomadicML is building a platform that unlocks machine intelligence from video footage. Their Vision-Language Models (VLMs) transform raw footage into structured intelligence for self-driving, robotics, and industrial automation. The founding team includes core contributors to ONNX Runtime and DeepSpeed.

Responsibilities

• Architect GPU inference pipelines for massive multi-GPU jobs on foundation-scale video models
• Build cloud upload infrastructure with integrations for AWS S3, GCP Storage, and Azure Blobs
• Design event-driven, autoscaling distributed processing pipelines using Kubernetes, Pub/Sub, or Ray
• Develop developer SDKs and APIs, including the NomadicML Python SDK
• Implement end-to-end observability for GPU utilization, job latency, and inference health
• Support web app integrations through backend endpoints and TypeScript SDK bindings

Requirements

• Deep proficiency in Python, Go, or TypeScript
• Experience with AWS, GCP, or Azure (IAM, S3/Blob Storage, Batch/Compute APIs)
• Strong understanding of GPU inference scaling, Kubernetes, and container orchestration
• Experience designing REST/gRPC APIs, SDKs, or developer-facing infrastructure
• Familiarity with asynchronous job orchestration such as Ray, Airflow, Dagster, or Temporal

Preferred Qualifications

• Experience with inference orchestration frameworks like DeepSpeed, Triton, or Ray Serve
• Understanding of video encoding, chunking, and streaming formats
• Basic front-end experience with React or Next.js
• Background in ML infrastructure, observability, or data management systems