Job Description

About Us

We're a stealth-mode startup backed by Team8 and founded by AI industry leaders. We are building the full-lifecycle trust platform for AI agents — delivering world-class evaluation, guardrails, synthetic data, and continuous optimization. We're at the forefront of a new engineering discipline, and we're building an exceptional team to match.

About the Role

Some engineers see a wall of limitations. You see the seam to slip through. We're looking for a senior infra engineer who moves fast and clears the path for the whole team. We want fast runners — people who get to "it works" while others are still scoping it. You're the builder who knows the rules cold, which is exactly why you know how to bend them safely. Whatever's blocking the team — infra that needs standing up, GPUs that need wrangling, an environment nobody's deployed to yet — you're the one who makes it disappear. You optimize for momentum without ever leaving a mess for tomorrow. Fast, fearless, and surgical.

In This Role You'll

• Stand up infra for POCs and experiments at startup speed — find the fastest safe path to "it works well"
• Wrangle GPUs and own LLM deployment & serving environments
• Build and own the testing infrastructure that lets the whole team move fast without breaking trust
• Know the rules well enough to bend them safely — ship velocity today without becoming tomorrow's incident
• Manage cloud environments and Kubernetes clusters across the stack
• Be the unblocker-in-chief: when something's in the way, you make it disappear

Requirements

What You Bring

Must have:

• Strong infra/platform background in cloud-native environments
• Deep Kubernetes expertise
• Hands-on with GPU infrastructure and LLM deployment/serving (e.g., vLLM, Triton, or similar)
• Strong cloud chops (AWS / GCP)
• A serious bias for action — you optimize for momentum and know exactly which corners are safe to cut
• Testing & CI infrastructure experience
• Comfortable scripting in Python / Bash (TypeScript a plus)

Bonus points:

• Terraform/IaC and GitOps
• Thrives in fast-moving 0→1 startup environments
• Security instincts sharp enough that "fast" never means "reckless"