Infrastructure Engineer / Early Stage Seed A / AI Research / Onsite at TrustIn
This startup is backed by $10M in seed funding and is scaling fast and looking for an infrastructure engineer to help own the systems at the core of it all.
WHAT YOU'LL DO:
• Own the dispatch layer for hundreds of concurrent agent runs, each with its own stateful environment. SQS, concurrency control, failure handling — you'll design and run it all at 100M tokens per minute across the fleet.
• Build evaluation environments that feel real and design tasks that genuinely push agents to their limits — not just what's easy to measure, but what actually matters.
• The agent evaluation space is moving fast. You'll support new environment modalities and ship integrations with external orchestration frameworks as the landscape evolves.
WHAT WE'RE LOOKING FOR:
✓3+ years building and scaling production distributed infrastructure — container orchestration, message queues, high-QPS systems
✓Extensive hands-on Kubernetes experience — you've run it under real load, not just set it up
✓Proven track record running LLM or large-scale ML workloads in production
✓Experience at both an early-stage startup and a fast-paced big tech company — you know how to build scrappily and at scale
✓Comfortable with message-driven architectures — SQS, Kafka, or similar