Staff Product Manager, Compute Platform

Open positions

Job Application for Staff Product Manager, Compute Platform at DigitalOcean


Back to jobs

New

Staff Product Manager, Compute Platform

Austin

Dive in and do the best work of your career at DigitalOcean. Journey alongside a strong community of top talent who are relentless in their drive to build the simplest scalable cloud. If you have a growth mindset, naturally like to think big and bold, and are energized by the fast-paced environment of a true industry disruptor, you’ll find your place here.  We value winning together—while learning, having fun, and making a profound difference for the dreamers and builders in the world.

We are looking for a Staff Product Manager, Compute Platform who is passionate about running managed compute services at scale, namely App Platform, Serverless Functions, and Container Registry, and who understands that a great platform experience is built on well-run infrastructure underneath it.

As a Staff Product Manager, Compute Platform at DigitalOcean, you will join a dynamic team dedicated to revolutionizing cloud computing. You will own the product strategy and roadmap for App Platform, Serverless Functions, and Container Registry (DOCR), the managed compute services that turn DO's Droplet fleet into a place where developers can ship without thinking about servers. That means owning both sides of each service, the developer-facing product and the platform economics beneath it, such as how workloads are packed onto compute, what each app, function, or image pull costs to serve, and how processor, virtualization, and container runtime trends shape our roadmap.

You will report to the Director of Product and partner day to day with the Hyderabad-based Compute Cloud Services engineering team that builds the business logic, APIs, and control plane for DO Compute, as well as with our Droplet/VPC, Kubernetes, and infrastructure product peers, finance, and GTM. The ideal person becomes the PM who can hold a customer conversation about deploy workflows in the morning and a capacity and margin conversation in the afternoon, the partner engineering trusts on architecture tradeoffs, and a core part of building out the compute platform PM group.

What you'll do

• Own the vision, strategy, and roadmap for App Platform, Serverless Functions, and DOCR, aligned with DO's overall compute platform strategy, and translate them into prioritized, well-scoped features that are simple to use and efficient to run.
• Own the platform economics of these services. Define how apps, functions, and registry storage map onto Droplets and shared compute, track cost to serve per unit of customer workload, and drive utilization, density, and bin-packing improvements with engineering and finance.
• Stay ahead of the compute landscape that underpins the platform, including processor roadmaps from Intel, AMD, and ARM, virtualization and microVM technologies, and container runtimes and isolation models, and translate those trends into platform decisions such as instance sizing, runtime choices, and price/performance tiers.
• Define the end-to-end developer journey across these services, from source repo or container image to a deployed, autoscaled, observable app or function, and remove the seams (auth, networking, registry, logs, billing) that make customers stitch it together themselves today.
• Work backwards from customers. Run discovery with developers, startups, and SMB platform teams, mine support and community signal, and turn it into clear jobs-to-be-done, user stories, and acceptance criteria.
• Own packaging and pricing for platform services, including how compute, egress, build minutes, storage, invocations, and image pulls are metered, how pricing tracks cost to serve, and how the services attach to Droplets, GPUs, managed databases, DOKS, and Spaces.
• Partner with engineering on architecture tradeoffs in the control plane and data plane, such as build pipelines, autoscaling and scheduling behavior, cold start and concurrency for Serverless Functions, registry performance and garbage collection, and multi-region placement.
• Drive performance, reliability, and efficiency outcomes with clear SLOs and cost targets, and champion the fixes that move them.
• Monitor the competitive landscape for managed compute and PaaS offerings, and identify where DO can differentiate on simplicity, price/performance, or both.
• Lead launches end to end with product marketing, docs, sales, and support, and own the results after GA, including adoption, activation, retention, and margin.

Key metrics:

• Platform services revenue and attach rate (share of Droplet, GPU, database, and DOKS customers also running App Platform or Serverless Functions, and share pulling from DOCR)
• Infrastructure efficiency, measured as fleet utilization and density for the compute serving these services, and cost per unit of customer workload (per app instance, per invocation, per GB stored and pulled).
• Gross margin per service, and the gap between pricing and cost to serve as the fleet and processor mix evolve.
• Activation, retention, and expansion, measured as time to first deploy and as apps, functions, and registries active at 30/90/180 days.
• Reliability and performance against SLO, e.g. deploy and build success rate, control plane availability, Serverless Functions cold start p95, image pull latency.

What you'll add to DigitalOcean:

• 8+ years of product management experience, with a proven track record defining and launching managed cloud compute, PaaS, serverless, or container products, ideally where you owned both the customer experience and the unit economics.
• Strong working knowledge of PaaS, serverless, and container registry models, including buildpack-based deploys, function runtimes and triggers, concurrency and cold start tradeoffs, and OCI image build, storage, and distribution.
• Compute fluency. A solid understanding of the cloud compute landscape, including CPU architectures (Intel, AMD, ARM) and their price/performance characteristics, virtualization and microVMs, container runtimes and isolation, and how these choices affect density, latency, and cost. You will not own the hardware roadmap,...