Technical Product Manager, Infrastructure

Vast.ai•Los Angeles, CA

1d•Onsite

About The Position

Vast.ai is seeking a Technical Product Manager to drive the backend our GPU cloud marketplace runs on. This is the software behind every live GPU rental (over 700k transactions a month): the daemon on every host's machines, the orchestration that powers each instance, and the infrastructure and test systems that let us ship it all at high velocity. The Scale: With over 20k GPUs, our AI cloud platform powers thousands of bleeding-edge training runs and critical production workloads for 120k+ developers all over the planet. This is a product with real scale, real data, and real users from day one. The work you ship moves tangible revenue in weeks, not quarters. The Challenge: We don't own the GPUs. Thousands of independent hosts price and operate them, and no two machines are alike. Your job is to make that heterogeneous, decentralized supply behave like the top-tier cloud our customers expect: fast, secure, and reliable.

Requirements

3+ years as a backend engineer AND 3+ years in product management
Hands-on experience with several of: security, scalability, reliability at scale, infrastructure tooling, distributed systems, observability, abuse/fraud prevention, compliance
Industry experience in one or more of the following areas: software infrastructure, developer tooling (APIs, SDKs, CLIs), AI/ML, cloud computing, GPUs, or two-sided marketplaces
Experience at a fast-paced startup or rapid-growth team
AI-native. You work with AI agents every day, and you want to build the compute layer the AI era runs on.
Deeply technical. You came up as an engineer, and you reason about scalability, security, and efficiency as product concerns, not afterthoughts.
Persuasive across the stack. You can make a technical tradeoff legible to c-suite and a thankless migration compelling to the team doing it.
Biased toward action. You'd rather ship the fix today than present the plan next week.

Responsibilities

Drive the infrastructure, security, and reliability roadmap: sequencing competing priorities, turning non-functional requirements into specs engineering can build, and seeing them through until they ship.
Own the metrics that prove it worked: platform uptime, fleet reliability, cost-to-serve.
Focus on Scalability & performance: Keep the core systems fast and ahead of demand as the marketplace grows, from database performance to end-to-end latency.
Focus on Security, trust & compliance: Harden the platform against attacks and abuse, and build the compliance roadmap enterprise customers need.
Focus on Observability & infra tooling: Give every engineer a clear view of platform health, and every host a clear view of their fleet. Own the metrics, tracing, and tooling behind it.