Product Manager, AI Infrastructure

Together AI Together AI · Data AI · San Francisco, CA · Product

Product Manager for AI Infrastructure at Together AI, focusing on GPU clusters, managed storage, and observability. The role involves owning day-to-day product work, identifying and resolving issues, running experiments, and translating customer needs into product decisions. The goal is to grow into full ownership of a product area within nine months. Requires strong technical foundation, data/analytics skills, and customer empathy for AI-native startups.

What you'd actually do

  1. Own the day-to-day product and feature work across Together's AI infrastructure products - GPU Clusters, Managed Storage, and observability - with an early focus on observability and storage.
  2. Find and drive problems to resolution with minimal guidance, including issues that aren't yet on anyone's radar.
  3. Run structured, hypothesis-driven experiments - reading the data yourself and driving the data collection and instrumentation needed to answer open questions.
  4. Partner across engineering, product, and partner teams to ship improvements and unblock work without waiting for permission.
  5. Translate a deep understanding of customers operating in fast-moving, high-ambiguity markets into product decisions.

Skills

Required

  • Deep empathy for customers operating in high-ambiguity, fast-evolving markets
  • Strong, hands-on data and analytics skills with a hypothesis-driven approach
  • High agency: you find problems, push through resistance, and drive resolution without waiting to be told what to do.
  • A real technical foundation -whether from an AI/infrastructure background, hands-on software development, or strong working knowledge of cloud and infrastructure.
  • A track record of collaborating across orgs and product lines to get things done.
  • Comfortable managing multiple priorities and workstreams at once.

Nice to have

  • Familiarity with the AI stack, especially at the infrastructure layer, is a strong plus.
  • Cloud platform experience (AWS, Azure, GCP, or similar) - with extra weight if you've helped _build_ such products.
  • working knowledge of the Nvidia/AMD GPU stack (InfiniBand, NCCL, GPU operator, etc.), or experience with AI training or inference workloads.

What the JD emphasized

  • AI infrastructure products
  • observability
  • GPU Clusters
  • Managed Storage
  • technical foundation
  • data and analytics skills
  • High agency