Senior Agent Infrastructure Engineer

About the role

Kairon’s runtime executes more than 12 million agent tasks a month. Each one is a chain of planning steps, tool calls, approvals and checks that has to be fast, cheap and fully traceable. You’ll join the four-person Runtime team that owns all of it.

This is a hands-on role. You’ll design and ship the systems that schedule agent steps, isolate tool execution and store traces, and you’ll be on call for them with the rest of the team.

“The best infrastructure work here is invisible to customers. They just notice their agents got faster and never lost a step.”

What you’ll do

• Own the step scheduler that runs agent plans with retries, timeouts and human approvals.
• Build sandboxed execution for tool calls across 400+ integrations, with per-tenant limits.
• Scale the trace store so every step stays searchable for 90 days at under 50ms p95.
• Cut cost per agent step through smarter caching, batching and model routing.
• Write design docs, review code and mentor engineers across the platform group.

What you’ll bring

• 6+ years building and operating distributed systems in production.
• Deep experience with Go, Rust or a similar systems language.
• A track record with queues, workflow engines or schedulers at high throughput.
• Comfort owning on-call and running calm, blameless incident reviews.
• Clear written communication. Most of our decisions start as a short doc.

Nice to have

• Experience running LLM workloads or evaluation pipelines in production.
• Background in multi-tenant isolation, sandboxing or WebAssembly.
• Open-source contributions to infrastructure projects.

Benefits

Competitive salary and meaningful equity, 30 paid days off, full health cover, 20 weeks of parental leave and a $2,000 yearly learning budget. Read about our values and team.