Senior AI Infrastructure Engineer Scale Secure Agent Runtimes

Socket.dev; San Francisco (CA); On-site; USD 120,000 - 190,000; Full time

Benefits offered by this job

Health insurance

401(k) plan

Team meals in office

Unlimited PTO

Job summary

Echelon is building an AI platform for Business Operations, working at high speed with founders and customers. The role focuses on secure execution substrates for agents, sandbox lifecycles, and scalable control-plane services in a multi-tenant environment.

You will own sandboxed code execution, observability, and durable job design, collaborating closely with teams to reduce latency and improve reliability. This is a hands-on infrastructure leadership position.

Qualifications

• 5+ years building production infrastructure or distributed systems.
• Hands-on ownership of multi-tenant workloads in production.
• Strong Linux knowledge across processes, filesystems, networking, and performance.

Responsibilities

• Design and operate sandbox lifecycle systems for agent code execution and browser work.
• Improve sandbox startup time, density, scheduling, and resource utilization.
• Build fast filesystem primitives for ephemeral and persistent agent state.
• Enforce tenant isolation, network policy, secrets boundaries, quotas, and least privilege access.
• Make long-running agent jobs durable via checkpointing and retries.
• Scale orchestration and control-plane services for bursty workloads.
• Build observability for resource pressure, failures, queue health, and latency.
• Run load tests, capacity plans, and incident reviews; fix root causes.

Skills

Production infra

Distributed systems

Developer platforms

Execution runtimes

Linux

Kubernetes

Infrastructure as Code

AWS

Azure

Go

Rust

TypeScript

Python

Tools

Firecracker

gVisor

Kata Containers

seccomp

eBPF

Job description

Echelon is building an AI platform for Business Operations, working at high speed with founders and customers. The role focuses on secure execution substrates for agents, sandbox lifecycles, and scalable control-plane services in a multi-tenant environment.

You will own sandboxed code execution, observability, and durable job design, collaborating closely with teams to reduce latency and improve reliability. This is a hands-on infrastructure leadership position.