Job Description

A stealth-stage AI infrastructure startup is building a self-healing system for software that automates defect resolution and development. Our platform is used by engineering and support teams to:

• Autonomously debug problems in software (technical support)

• Fix issues directly in code

• Prevent these problems from recurring

The company is backed by leading investors including Foundation Capital, WndrCo, and Green Bay Ventures — along with prominent operators such as Matei Zaharia, Drew Houston, Dylan Field, Guillermo Rauch, and others.

We believe that as software development accelerates, engineering and support teams face mounting challenges maintaining quality and reliability. We see this as a rare opportunity to reinvent how modern software is supported — with intelligent, automated infrastructure.

About the Role

We’re searching for a Platform Engineer who lives and breathes distributed systems You’ll help build the foundation that synchronizes petabytes of data, indexes billions of lines of code, and coordinates fleets of AI agents working in parallel. If terms like “five-nines,” “exact-once,” or “sub-second latency” light you up — you’ll thrive here.

In this role, you will:

• Design cloud-native architectures that elastically scale to support thousands of microservices and GPU workers.

• Build high-throughput data planes for log ingestion, change-data-capture (CDC), and real-time feature stores powering our self-healing system.

• Implement ultra-low-latency indexes (vector, inverted, graph) to support semantic queries across massive codebases.

• Synchronize state across clusters and regions for global enterprise users — ensuring consistent, fast results regardless of data scale.

• Automate LLM-agent orchestration so hundreds of autonomous fix bots can work concurrently without collisions.

• Harden reliability and security with chaos engineering, live migrations, and deep protections for customer data.

• Partner closely with ML researchers and product leads to translate novel ideas into hardened infrastructure.

You might thrive in this role if you:

• Have 5–10+ years of experience building and maintaining distributed systems in production.

• Understand consensus protocols, sharding, replication, and backpressure inside and out.

• Have successfully managed high-volume, multi-tenant data pipelines in real-time environments.

• Are comfortable navigating and debugging massive codebases and infra at scale.

• Write infra-as-code (Terraform / Pulumi) and automate with tools like Argo, GitHub Actions, or Buildkite.

• Take full ownership: you architect it, you build it, you run it.

Let me know if you'd like this version made more company-facing (for job boards) or stealth investor-facing (for fundraising decks).