About the Role
Focus on Sailboxes, our cost-efficient CPU sandbox for long-horizon agents. You’ll be one of the first engineers on a rapidly growing product-line and write code at every part of the agent runtime, from low-level performance work on our custom networking stack to building large scale systems that maximize our efficiency.
Responsibilities
- Extract maximum performance from the platform by digging into the custom FirecrackerVM fork.
- Model historic usage patterns and implement scheduling algorithms to maximize fleet utilization and autoscale intelligently while minimizing latency.
- Design and implement highly-available distributed systems to handle millions of operations per second.
- Build deep monitoring tooling and features to make Sailboxes a delightful experience for humans and agents.
- Own future product directions for Sailboxes, including Windows/Mac sandboxes and high-efficiency storage.
Requirements
- Strong distributed systems fundamentals, including concurrency, networking, databases, and performance engineering.
- Experience building and maintaining software for large-scale production distributed systems.
- Experience with cloud infrastructure (GCP, AWS, Azure), container orchestration, and/or multi-cloud networking.
- A product-focused approach to platform work, with a commitment to building robust, scalable, and ergonomic solutions.
- Experience with FirecrackerVMs is a plus.
Skills
- Distributed systems
- Concurrency
- Networking
- Databases
- Performance engineering
- Cloud infrastructure (GCP, AWS, Azure)
- Container orchestration
- Multi-cloud networking
- FirecrackerVMs
Location
- San Francisco
Work Type
- Onsite
Experience Level
- Early career engineer on a rapidly growing product-line
Benefits
- All meals provided
- Studio Display at each desk
- Investment in time-saving tools
- Multiple coffee/tea options
About the Company
- Builds the world's most efficient software for inference and agent hosting.
- Technologies allow customers to deploy AI agents at large scale.
