Agentic Infrastructure Engineer — Frontier AI Lab (Confidential) | AIONIA


Confidential Role

Agentic Infrastructure Engineer — Frontier AI Lab

Building execution-layer infrastructure for intelligent systems at organizations operating at the boundary of what AI can do.
San Francisco or New York
$250K – $500K
Agent Runtime
Distributed Systems
Python
Frontier AI

Apply for this role

About the role

Sourcing engineers for a small number of confidential roles at frontier AI labs — organizations building at the boundary of what intelligent systems can do.

These are not application-layer roles. The work is at the execution layer: the runtime infrastructure that determines how agents behave, fail, recover, and improve in production.

What the work looks like

Building and owning agent harnesses and orchestration frameworks. Designing execution logic for multi-model, multi-tool systems. Creating sandboxed environments where agents can act safely. Building eval and observability systems that close the feedback loop.

Operating at the intersection of systems reliability and model behavior.

Who we’re looking for

You have deep backend or distributed systems experience — fintech, infra, or high-reliability production environments preferred. You’ve owned systems end-to-end, from design through production and iteration. You think about failure modes before you think about features.

You may not have “AI” in your current title, but you’ve been building toward this problem.
Must-haves
  • Deep backend or distributed systems experience — fintech, infra, or high-reliability production environments strongly preferred
  • Owned complex systems end-to-end from design through production, iteration, and long-term maintenance
  • Thinks about failure modes, recovery paths, and observability before features
  • Strong Python proficiency with genuine production-level systems work
  • Must work on-site in San Francisco or New York full-time
  • No visa sponsorship available
Nice to have
  • Prior experience at an AI lab, research org, or safety-adjacent engineering team
  • Familiarity with agent frameworks, LLM orchestration, or eval infrastructure
  • Experience building sandboxed or isolated execution environments
  • Background in high-reliability systems — fintech, infrastructure, or defense-adjacent
  • Interest in the long-term trajectory of AI systems and safe deployment at scale
Deal breakers — do not apply if
  • Your experience is primarily at the application or product layer
  • You haven’t owned complex production systems end-to-end with real operational stakes
  • You think about features before failure modes
  • You’re looking for someone else to define the system architecture
  • You require visa sponsorship
  • You cannot work on-site in San Francisco or New York full-time

Compensation

$250,000 – $500,000+ total comp depending on level and organization. Equity at organizations of this caliber is meaningful, not nominal.

Interview process

1
Confidential intro call
Background, interests, role fit, no commitment required

2
Lab introduction
AIONIA introduces you to the relevant organization if there’s a match

3
Technical and systems interview
Engineering depth, systems design, problem-solving

4
Onsite / final loop
Full interview loop with the team, format varies by organization

All conversations are confidential. Reach out even if you’re not actively looking.

Apply for this role