We are looking for a Founding AI Infrastructure Engineer to build Pluto’s compute platform from the ground up.
This person will own the technical backbone of Pluto’s compute business: customer-facing APIs, provider integrations, provisioning workflows, cluster lifecycle management, billing/metering infrastructure, and the internal tools needed to scale beyond manual brokerage.
You will work directly with the founders to transform messy, real-world compute procurement into clean software abstractions. The goal is to let customers request, launch, manage, and pay for GPU clusters through Pluto, while Pluto handles routing, provisioning, and orchestration across underlying neoclouds.
This is a deeply technical, high-ownership role for someone who wants to build core infrastructure at the intersection of AI, cloud, marketplaces, and financial infrastructure.
What You’ll Do
Design and build Pluto’s customer-facing compute API for provisioning and managing GPU clusters
Build control-plane systems for cluster orchestration, lifecycle management, and state tracking
Integrate with neoclouds, GPU providers, and infrastructure partners through APIs, Kubernetes, VMs, bare metal, and manual/semi-automated workflows
Create provider abstraction layers that normalize fragmented infrastructure into a unified Pluto resource model
Build internal tooling to replace manual brokerage workflows with repeatable, scalable software systems
Develop metering and usage-tracking systems for GPU-hours, storage, networking, and customer billing
Build systems for capacity discovery, quoting, reservations, and provider routing
Work closely with customers to understand how AI teams want to launch, manage, and scale compute
Improve reliability, observability, and operational visibility across provisioned clusters
Define clean interfaces between Pluto, customers, and underlying compute providers
Write production-quality code and make foundational architecture decisions for the platform
What We’re Looking For
Strong backend, infrastructure, or platform engineering background
Experience building APIs, control planes, orchestration systems, or infrastructure automation
Strong systems fundamentals across Linux, networking, storage, and distributed systems
Experience with Kubernetes, VMs, containers, or bare-metal infrastructure
Ability to build reliable services in Python, Go, Rust, TypeScript, or similar languages
Comfort working with cloud APIs, infrastructure-as-code tools, schedulers, and provisioning workflows
Ability to turn messy operational processes into clean product abstractions
Strong product sense around APIs, defaults, developer experience, reliability, and operational simplicity
High agency and comfort operating in an early-stage startup environment
Excellent written communication and ability to document technical decisions clearly
Nice to Have
Hands-on experience with GPU infrastructure or AI workloads
Experience with H100, H200, B200, A100, L40S, or similar GPU environments
Experience building systems around Kubernetes, Slurm, Ray, or other cluster managers
Experience with multi-tenant infrastructure platforms
Experience building billing, metering, quota, or usage-based pricing systems
Experience at a cloud provider, neocloud, AI infrastructure company, or high-growth developer infrastructure startup
Familiarity with AI training, inference, model-serving, or distributed workloads
Experience designing APIs, CLIs, SDKs, or developer-facing infrastructure products
Why Join Pluto
Compute is becoming the core commodity of the AI economy, but the market is still extremely fragmented. AI companies need GPUs, neoclouds need distribution, and there is no clean software layer connecting demand to supply.
Pluto is building that layer.
You will have the opportunity to design and build the first version of Pluto’s compute platform, work directly with customers and providers, and turn a manual brokerage business into a scalable infrastructure marketplace.
This is a role for someone who wants to build foundational systems, not just features. You will help define how AI companies access compute and how GPU capacity becomes a programmable, liquid market.