Description

You own the build of QE enablement as a GenRuntime platform capability, shipping production code weekly. You build agents and platform hooks that automatically emit test assets and quality gates for every app scaffolded through GenOS.

This role is on-site.

Responsibilities

• Build QE-generation agents on GenRuntime to consume dev-flow artifacts and emit validated functional, API, and regression test assets.
• Wire QE enablement into the paved road for self-serve onboarding with zero hand-holding.
• Develop reusable GenRuntime tools for test execution, self-healing selectors, and failure triage.
• Extend AI Workbench eval primitives into release gates, enforcing statistical quality thresholds in CI/CD.
• Embed safety, privacy, and moderation regressions into generated tests and integrate with Feature Management.

Required Skills

• 7+ years backend/platform engineering in Java, Python, or Go with deep distributed-systems fundamentals.
• 2–3+ years building LLM-powered systems in production using LangChain, LlamaIndex, or homegrown frameworks.
• Strong structured-output engineering skills including schema enforcement and output validation.
• Proven QE domain depth in test automation architecture, CI quality gates, and flaky-test economics.
• Experience with a major LLM provider at scale, such as AWS Bedrock, handling rate limits and latency variance.
• Deep AWS and Kubernetes deployment experience, specifically on Intuit Kubernetes Service.
• Fluency in AI-assisted development workflows using Copilot-class tooling.

Preferred Skills

• LLM evaluation engineering experience with golden datasets and LLM-as-judge calibration.
• Experience with vector stores and retrieval in production (OpenSearch, pgvector, or Pinecone).
• Background in multi-tenant enterprise platforms under strict security/compliance.

Key Skills

Education

Any Graduate