Lead Test Framework Architect (M/F/D)

Agilent is building a Kubernetes-based software platform spanning instruments, applications, and shared services across multiple product lines. We are embedding AI into how we build and verify software. We are looking for a Lead Test Framework Architect to lead the adoption of AI in our software test lifecycle, and to build the test frameworks, evaluation methods, and governance that let us ship AI-accelerated software safely in regulated, GxP / 21 CFR Part 11 environments.
This is a hands-on senior individual-contributor role. AI accelerates software delivery, but that acceleration exposes weaknesses downstream without a strong automated-testing control system. Building that control system - and leading the organization to adopt it - is the core of this role.

• →Lead AI SDLC adoption for quality engineering. Own the strategy for AI-assisted test authoring, execution, triage, and maintenance. Prove value through reference workflows, then scale adoption across product teams.
• →Build AI-native test frameworks. Design workflows that generate unit, contract, integration, and E2E suites from specifications and intent. Verify that AI-generated tests exercise specified intent, not implementation. Integrate GitHub Copilot (enterprise standard) and complementary tools (e.g., Qodo, Diffblue, Playwright AI agents, self-healing frameworks) into CI/CD quality gates.
• →Test our AI features. Build evaluation and regression frameworks for LLM- and agent-powered product capabilities: non-deterministic evaluation, hallucination and grounding scoring, drift detection, and AI observability using OpenTelemetry GenAI conventions.
• →Own the regulatory bridge. Integrate AI-assisted testing into validated environments using risk-based Computer Software Assurance (CSA), GAMP 5 (2nd Edition) and the ISPE GAMP AI Guide, and - for diagnostics software - IEC 62304. Preserve audit trails, traceability, and data integrity for AI-generated artifacts.
• →Establish human-in-the-loop governance. Define risk-scaled review checkpoints, acceptance criteria, and rollback triggers for AI-generated code and tests.
• →Define platform-wide test architecture standards for a Kubernetes-based microservices platform: testability-by-design, test environment strategy (ephemeral namespaces, environment-as-code), CI/CD quality gate architecture, contract testing for microservices and APIs, test data strategy (synthetic data, isolation, AI/ML data requirements), and E2E frameworks that work reliably in containerized environments.
• →Instrument and report on impact. DORA delivery metrics augmented with AI-specific measures (coverage delta, defect-escape rate, mean-time-to-triage, eval-suite pass rates). Lead the change effort - champions, enablement, playbooks - that drives durable adoption.

• 10+ years of software engineering, with 3+ years in test architecture or quality engineering architecture roles.
• Demonstrated daily, hands-on use of AI-assisted development and testing tools (GitHub Copilot, Claude Code, Cursor, or equivalent) with the ability to speak concretely to effective usage patterns and failure modes.
• Deep hands-on Kubernetes expertise for test environment design: ephemeral namespace provisioning, container-based test infrastructure, environment-as-code.
• Strong experience designing multi-level test automation frameworks: unit, contract, integration, E2E, and performance testing in distributed systems.
• CI/CD pipeline architecture and quality gate integration (GitHub Actions, Jenkins, or equivalent).
• Contract testing experience (Pact or consumer-driven contract testing) for microservices and API-first architectures.
• Demonstrated ability to deliver reusable test frameworks adopted across multiple product teams — reference implementations, not just guidelines.
• Working knowledge of testability-by-design and the ability to influence how software is architected to be inherently more testable.
• Experience leading adoption and change in an established engineering organization, with a track record of defensible outcome metrics.

• Experience testing AI/ML, LLM, or agent-based systems: evaluation frameworks (deepeval, promptfoo, LangSmith, Langfuse, Maxim AI, or equivalent), non-deterministic regression, hallucination and grounding scoring, drift detection.
• Fluency in regulated software validation: 21 CFR Part 11, GxP / GAMP 5, CSA vs. CSV, and IEC 62304 for medical device or diagnostics software.
• Context engineering for LLMs: prompt/spec design, RAG, MCP servers, project-memory patterns.
• Observability tooling in containerized environments, ideally with OpenTelemetry GenAI Semantic Conventions.
• Performance and load testing at platform scale (k6, Gatling, Locust, or equivalent).
• Infrastructure-as-code and GitOps applied to test environment provisioning.
• Background in scientific instrument software, laboratory platforms, or life sciences / diagnostics.