Staff Engineer - Agentic Data Sources and Harness

About the Role

Our AI agent runs in production at some of the world's largest companies — from airlines to sports betting to fintech — investigating downtime, querying petabytes of data, and opening pull requests to solve production issues.

We're hiring a Backend Engineer to make this agent smarter, faster, and more capable.

Why This Role Is Different

We're AI-first — not just in what we build, but in how we build it.

The process for developing a new feature looks different here. Our engineers regularly join customer calls and hear feature requests directly. The next step is spinning up coding agents in sandbox environments to start building — often while still on the same call.

The agent then tests the feature against isolated cloud accounts, iterating until it works. It submits the code for an automated review from CodeRabbit, addresses the review comments, and iterates again. Only after passing both automated tests and AI code review does a human look at the code for the first time.

This enables a velocity that most engineers aren't used to. We often ship fixes to customers on the same call. Because of the investments we've made in automated verification, we can do so without compromising on quality.

We're not just building software — we're building an operating system for building software, and pushing the limits of what AI can do.

What You'll Do

• Join calls with SRE teams at Fortune 500 companies, understand their incident workflows, and go build what they need
• Optimize agent accuracy, latency, and reliability through context engineering, subagents, and evaluation frameworks
• Design observability for the agent itself — tracing its reasoning and monitoring token usage
• Architect systems that enable AI agents to develop, test, and verify features autonomously — designing the right abstractions so this works reliably at scale
• Solve hard problems around agent accuracy and safety — how do you verify an agent's answer when there's no ground truth, and how do you let it take actions without breaking things?

What We're Looking For

• 8+ years of backend development experience
• Strong Python and backend fundamentals — the better your architectural instincts, the more effectively you can leverage AI to build at speed
• Strong debugging skills and systems thinking — you reason about complex systems, not just individual components
• Comfortable in a fast-moving startup where you own problems end-to-end with high autonomy
• Strong English communication skills

Bonus Points

• Hands-on experience building AI agents using OpenAI or Anthropic APIs
• Built MCP servers or similar tool integration systems
• Kubernetes experience — deploying, debugging, and managing production workloads
• Observability/monitoring tools — Prometheus, Grafana, Datadog, New Relic, Loki, or similar
• Experience building developer tools or CLI applications
• Open source contributions (especially CNCF projects)
• DevOps/CI/CD experience

About Us

We have an office in Herzliya. The role is hybrid, with the option to work fully remote or come into the office.