Staff Software Engineer

Harrison Clarke; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Harrison Clarke in San Francisco is seeking a hands-on systems engineer to help build the platform that powers AI agents. You will own the core systems that orchestrate agents, tool execution, state management and failure recovery, ensuring reliability and scalability in production.

You will design context management, memory, file storage and sandboxed execution, own model routing and serving, and manage Kubernetes and cloud infrastructure to make long-running workloads faster, safer and cheaper

Qualifications

• Deep CS fundamentals and experience building large-scale production systems.
• Ability to debug anywhere from app code to runtimes and networks.
• US work authorization.

Responsibilities

• Build the system that runs the agents: orchestration, tool execution, state, and failure recovery.
• Develop context management, memory, file storage, and sandboxed execution.
• Own model routing, serving, Kubernetes, networking, and cloud infrastructure.
• Improve long-running agent workloads to be faster, safer and cheaper to run.

Skills

Deep computer science fundamentals

Debug across stack

Authorization to work in the US

Tools

Kubernetes

Cloud infrastructure

Job description

Our client is building the platform companies use to run their business on AI agents. A customer describes the outcome they want in plain English, and agents get the work done across their tools, either right away or running in the cloud 24/7. Revenue has grown more than 10x this year.

The founder co-founded one of the most widely used developer platforms in the world, which was acquired by a FAANG company & successfully integrated into their platform. He's a repeat founder and an active angel investor.

Why this role

• Backed by Tier 1 VCs and well-known angels from AI and fintech
• A hands-on IC role where everyone, including the most senior people, builds and ships
• You'd own the core systems that make the agents capable, reliable and scalable

What you'll do

• Build the system that runs the agents: orchestration, tool execution, state, evaluation and recovery from failures
• Build context management, memory, file storage and sandboxed execution
• Own model routing and serving, Kubernetes, networking and cloud infrastructure
• Make long-running agent workloads faster, safer and cheaper to run

What you bring

• Deep computer science and systems fundamentals, and a track record of building and running complex production systems
• The ability to debug anywhere in the stack, from application code down through runtimes, storage, networking and infrastructure
• Authorization to work in the US