Staff Product Manager - Observability

Lambda•Bellevue, WA

1d•$291,000 - $430,000•Hybrid

About The Position

Lambda is hiring a Staff Product Manager to own observability as a product across its AI cloud. The role covers the customer-facing experience—including GPU and cluster health, utilization, job-level telemetry, and InfiniBand fabric metrics—as well as the internal platform telemetry layer that supports fleet operations and reliability. The product manager will work closely with SRE, fleet engineering, and console and API teams to create a coherent observability experience for customers and operators, from individual on-demand GPU instances to clusters with 64 to 1,024+ GPUs. This position reports to the Head of Platform Product Management. The role calls for rigorous, evidence-based product judgment, the ability to influence teams without formal authority, and end-to-end execution from product idea through launch, measurement, and iteration. Deep observability domain experience is a strong plus, while insight, influence, and execution are the core expectations. The position requires working from Lambda’s Bellevue or San Francisco office four days per week; Tuesday is currently the designated work-from-home day.

Requirements

At least 7 years of product management experience, including at least 3 years working on technical infrastructure, platform, or developer-facing products.
Demonstrated ability to turn data and customer feedback into product decisions, with examples of insights that changed a roadmap.
Ability to influence engineers, designers, executives, and customers without formal authority; experience shipping products that required cross-team adoption.
Experience taking products from idea through launch, measurement, and iteration, and articulating lessons learned from post-launch metrics.
Comfort with deeply technical discussions about distributed systems, including metrics pipelines, hardware health, and failure modes.
Strong, concise writing skills, including the ability to align stakeholders with a one-page document.
Ability to work effectively amid ambiguity and take ownership of a domain as its first product manager.
Ability to define iterative plans that move an organization from its current state toward a desired outcome.

Nice To Haves

Experience shipping monitoring or observability products, such as products in the Datadog, Grafana, or Prometheus class.
Experience with HPC or distributed-systems telemetry, including interconnect and fabric-level metrics.
Experience building developer tools or API-first products.
Hands-on exposure to distributed training stacks such as PyTorch with NCCL, or GPU fleet tooling such as DCGM.
Experience in a usage-based cloud infrastructure business.
Deep observability domain experience.

Responsibilities

Define and drive the observability product strategy and roadmap across Lambda’s cloud, from individual on-demand GPU instances to large GPU clusters.
Choose which telemetry signals customers see—including GPU and cluster health, utilization, job-level telemetry, and InfiniBand fabric metrics—and how they are presented through the console and Lambda Cloud API.
Build diagnostic experiences that help customers distinguish issues in their code or configuration from platform problems such as a failing node or degraded InfiniBand link.
Work with each product to define and productize SLIs, SLOs, and SLAs, including expectations and tolerances for platform performance.
Partner with SRE and fleet engineering to define internal telemetry for fleet operations and reliability, using a shared data foundation for customers and operators.
Engage AI research labs, AI startups, and enterprise machine-learning teams to understand how they debug distributed training and inference, and use those insights to guide product decisions.
Align engineers, designers, and executives around the roadmap through clear writing and evidence rather than formal authority.
Define success metrics for observability features, instrument adoption, and use results to guide subsequent product decisions.
Drive observability capabilities through launch across the console, Lambda Cloud API, and cluster tooling.