What if the work you did every day could impact the lives of people you know? Or all of humanity?
At Illumina, we are expanding access to genomic technology to realize health equity for billions of people around the world. Our efforts enable life-changing discoveries that are transforming human health through the early detection and diagnosis of diseases and new treatment options for patients.
Working at Illumina means being part of something bigger than yourself. Every person, in every role, has the opportunity to make a difference. Surrounded by extraordinary people, inspiring leaders, and world changing projects, you will do more and become more than you ever thought possible.
Location
San Diego, CA. Not remote.
Position Summary
We are hiring a Senior Staff AI Engineer to set the technical direction for quality engineering across Illumina's enterprise platforms, with AI agents doing a growing share of the testing work. You will be the most senior individual contributor in QA: you design the test architecture, lead our Tricentis platform, and make agentic testing safe enough for regulated, GxP-relevant systems. Technical leadership of 10-20 QA and automation engineers across vendors and teams.
The role sits in Enterprise AI and partners with Data Engineering, application teams (SAP, Salesforce, ServiceNow, Workday, M365) and Quality & Regulatory. It covers two equally important missions: using AI agents to test our software faster, and testing the AI agents we build so they can be trusted in production.
Responsibilities
- Own the enterprise test strategy. Define a risk-based, shift-left and shift-right strategy across SAP S/4HANA, Salesforce, ServiceNow, Workday, custom apps, data pipelines and AI agents.
- Lead the Tricentis platform. Architect Tosca, qTest, NeoLoad, SeaLights, LiveCompare and Tricentis AI Workspace as one quality platform, including standards, reuse libraries, licensing and roadmap.
- Build agentic testing at scale. Design and run multi-agent test workflows that generate, execute, heal and triage tests, with clear human approval points and audit trails.
- Test our AI agents. Create the evaluation framework that gates every agent release: golden datasets, LLM-as-judge scoring, red teaming, tool-call correctness, safety and cost checks.
- Connect agents to tools through MCP. Integrate Tricentis remote MCP servers and our own MCP servers with Claude, GitHub Copilot, Cursor and Codex under the Enterprise AI gateway and policy.
- Embed quality in CI/CD. Make test impact analysis, quality gates and release risk scoring standard in every pipeline, so teams ship on evidence rather than gut feel.
- Deliver validated quality for GxP. Partner with Quality & Regulatory to apply FDA CSA and GAMP 5 (2nd ed.) so automated and AI-assisted testing produces Part 11 compliant evidence.
- Raise the bar for engineers. Mentor senior and staff engineers, run design reviews, write the playbooks, and lead the QA community of practice.
- Measure and report value. Own the quality scorecard (escape rate, automation coverage, cycle time, cost per test) and report it to engineering and business leaders.
Experience & Education
- 12+ years in software quality engineering, with 5+ years as a lead, principal or staff level engineer setting strategy across multiple teams.
- 5+ years of hands-on Tricentis Tosca experience at enterprise scale, plus working knowledge of qTest and at least one of NeoLoad or SeaLights.
- Proven delivery of test automation for packaged enterprise apps, ideally SAP S/4HANA or ECC and Salesforce.
- 1+ year building or operating LLM-based or agentic testing in production, not only proofs of concept.
- Strong coding in at least two of Python, TypeScript/JavaScript, Java or C#.
- Deep CI/CD experience with GitHub Actions, Azure DevOps, GitLab or Jenkins, including quality gates.
- Experience testing in regulated environments (GxP, medical device, SOX or similar) and producing audit-ready evidence.
- Clear written and spoken communication with engineers, product owners and executives.
- Bachelor's degree in Computer Science, Engineering or a related field, or equivalent experience.
Preferred qualifications
- Tricentis certifications (Tosca Architect, qTest, NeoLoad); ISTQB Advanced or Expert, or ISTQB Certified Tester AI Testing (CT-AI).
- AWS Certified Machine Learning or AI Practitioner, or Databricks Generative AI Engineer certification.
- Published MCP servers, eval suites or open source test tooling.
- Experience in life sciences, genomics, diagnostics or medical devices.
- Experience leading an SAP S/4HANA migration or RISE program test effort.
Tricentis platform skills
We expect expert depth in Tosca and qTest, and practical command of Tricentis's agentic products released in 2025 and 2026.
- Tosca (Commander, Cloud, Vision AI, model-based test automation): Architect module and test case design, reusable libraries, test data service, distributed execution and Tosca Cloud migration
- Tosca Copilot and Agentic Test Automation: Generate full test cases from natural language and Vision AI, review output, and govern what agents may change
- qTest (Manager, Insights, Explorer, qTest Copilot, Agentic Test Creation): Run test management, traceability to requirements and Jira, and agent-assisted test authoring
- Tricentis AI Workspace: Onboard and orchestrate Tricentis and third-party agents with approvals, governance and human oversight
- Tricentis remote MCP servers: Connect Tosca, qTest and NeoLoad to Claude, Copilot, Cursor and our own agents through the Enterprise AI MCP gateway
- SeaLights (quality intelligence): Apply test impact analysis and code coverage to cut regression runs and find untested change
- NeoLoad and Agentic Performance Engineering: Design performance and load tests, including AI-generated user paths for SAP and web apps
- LiveCompare: Run SAP change impact analysis to scope regression for transports and upgrades
- Testim and Tricentis Device Cloud: Automate web and mobile UI tests with self-healing locators across real devices
- Tricentis Data Integrity: Validate data pipelines, migrations and reports end to end
- Newer Tricentis Labs tools (Aida, AgentScore, Release Risk Intelligence): Pilot autonomous exploration, agent quality scoring and release risk scoring, and advise on adoption
- Tricentis certifications (Tosca Automation Specialist and Architect, qTest, NeoLoad) are strongly preferred. Product names reflect Tricentis's portfolio as of October 2026 (Pulse 2.0 interview, VKTR, Business Wire).
AI agentic testing skills
You will be fluent on both sides: agents that test software, and testing of the agents themselves.
Agents that test software
- Designing multi-agent QA workflows (planner, generator, executor, healer, triage, reporter) with defined roles, hand-offs and stop conditions.
- Building agents with Claude Agent SDK, Amazon Bedrock AgentCore, LangGraph, CrewAI, Microsoft Copilot Studio or Databricks Agent Bricks.
- Model Context Protocol (MCP) servers and clients, and Agent2Agent (A2A) protocol for agent-to-agent hand-offs.
- AI-native browser and UI automation: Playwright MCP, Claude computer use, browser agents and visual AI (Tricentis Vision AI, Applitools).
- Agentic coding assistants for test code: Claude Code, GitHub Copilot agent mode, Cursor and Codex, with review and branch protection rules.
- Autonomous exploratory testing, self-healing locators, and AI-driven defect triage, clustering and root cause suggestions.
- Prompt and context engineering: skills, system prompts, retrieval of requirements and past defects, structured outputs.
- Synthetic test data generation with privacy controls (no real PHI or PII in prompts).
Testing AI agents and LLM applications
- Evaluation design: golden datasets, rubric scoring, LLM-as-judge with calibration against human labels, pass@k and regression baselines.
- Eval tooling such as Bedrock AgentCore Evaluations, MLflow 3 GenAI evaluation, DeepEval, Ragas, Promptfoo, LangSmith, Arize Phoenix or Braintrust.
- Agent-specific checks: tool-call correctness, trajectory and plan quality, memory behaviour, multi-turn consistency, latency and token cost.
- RAG quality: retrieval relevance, groundedness, citation accuracy and hallucination rate.
- Red teaming and security against the OWASP Top 10 for LLM and Agentic AI: prompt injection, tool misuse, data exfiltration, excessive agency (Garak, PyRIT or similar).
- Guardrail and policy testing (Bedrock Guardrails, Cedar-style policy, content filters) and human-in-the-loop checkpoints.
- Observability with OpenTelemetry GenAI conventions and trace-based test assertions.
- Risk-tiered release gates aligned to NIST AI RMF, ISO/IEC 42001 and the EU AI Act.
Modern engineering skills
- Code-based automation: Playwright, Selenium 4, Cypress, WebdriverIO; Appium for mobile; Pytest, JUnit 5, TestNG
- API and contract testing: Postman, REST Assured, Karate, Pact contract testing; GraphQL, gRPC and event (Kafka) testing
- Performance and reliability: NeoLoad, k6, JMeter, Gatling; chaos engineering (AWS FIS, Gremlin); SLO-based testing
- Security testing: SAST, DAST and SCA in pipelines (Snyk, Checkmarx, OWASP ZAP, Burp Suite); secrets scanning; SBOM checks
- Accessibility: WCAG 2.2 AA automation with axe-core and manual assistive technology checks
- Data and analytics testing: Databricks, Unity Catalog, Delta Live Tables / Lakeflow, dbt tests, Great Expectations; SQL at an expert level
- CI/CD and DevOps: GitHub Actions, Azure DevOps, Jenkins; trunk-based development, feature flags, canary releases; test impact analysis
- Cloud and containers: AWS (Bedrock, Lambda, ECS/EKS), Docker, Kubernetes, ephemeral test environments, Terraform
- Observability: OpenTelemetry, Datadog, Splunk or Grafana; production monitoring as test signal (shift-right)
- Service virtualization and test data: Tosca OSV, WireMock, mock servers; masked and synthetic data management
- Enterprise apps: SAP S/4HANA and Fiori, Salesforce, ServiceNow, Workday, Microsoft 365
Regulated quality and leadership
Regulated quality
- FDA Computer Software Assurance (CSA) and GAMP 5 Second Edition, including critical thinking and risk-based assurance in place of script-heavy validation.
- 21 CFR Part 11 and EU Annex 11: electronic records, signatures and audit trails for automated and AI-generated test evidence.
- ISO 13485 and IEC 62304 awareness for software that touches medical device or diagnostic workflows.
- FDA and EMA good AI practice principles for AI used in regulated processes.
- Data privacy in testing: HIPAA, GDPR and data residency for test data and AI prompts.
Leadership expectations at Senior Staff level
- Set multi-year technical direction and get it adopted without direct authority.
- Make build, buy and partner calls, including Tricentis versus open source trade-offs, with clear cost and risk reasoning.
- Lead vendor and system integrator partners to a single quality standard.
- Coach engineers into staff-level thinking and grow the bench of AI-fluent testers.
- Explain quality risk to executives in business terms and recommend ship or hold decisions.
What success looks like in the first 12 months
- Agentic test strategy and Tricentis reference architecture:
- Target: Approved within 90 days
- Regression automation coverage on tier 1 apps
- Regression cycle time for SAP releases
- AI agents in production behind an eval release gate
- Production defect escape rate
- Target: Down 30% from baseline
- GxP systems with CSA-aligned, Part 11 compliant automated evidence
- Target: All in-scope systems
#LI-HYBRID
The estimated base salary range for the Senior Staff AI Engineer, (AI Quality & Test Automation) role based in the United States of America is: $141,600 - $212,400. Should the level or location of the role change during the hiring process, the applicable base pay range may be updated accordingly. The range reflects long‑term growth in the role; therefore, most candidates are hired between the minimum and middle of the range. Placement depends on experience, skills, location, and internal equity. Additionally, all employees are eligible for one of our variable cash programs (bonus or commission) and eligible roles may receive equity as part of the compensation package. We offer a wide range of benefits as innovative as our work, including access to genomics sequencing, family planning, health/dental/vision, retirement benefits, and paid time off.
We are a company deeply rooted in belonging, promoting an inclusive environment where employees feel valued and empowered to contribute to our mission. Built on a strong foundation, Illumina has always prioritized openness, collaboration, and seeking alternative perspectives to propel innovation in genomics. We are proud to confirm a zero-net gap in pay, regardless of gender, ethnicity, or race. We also have several Employee Resource Groups (ERG) that deliver career development experiences, increase cultural awareness, and offer opportunities to engage in social responsibility. We are proud to be an equal opportunity employer committed to providing employment opportunity regardless of sex, race, creed, color, gender, religion, marital status, domestic partner status, age, national origin or ancestry, physical or mental disability, medical condition, sexual orientation, pregnancy, military or veteran status, citizenship status, and genetic information. Illumina conducts background checks on applicants for whom a conditional offer of employment has been made. Qualified applicants with arrest or conviction records will be considered for employment in accordance with applicable local, state, and federal laws. Background check results may potentially result in the withdrawal of a conditional offer of employment. The background check process and any decisions made as a result shall be made in accordance with all applicable local, state, and federal laws. Illumina prohibits the use of generative artificial intelligence (AI) in the application and interview process. If you require accommodation to complete the application or interview process, please contact accommodations@illumina.com. To learn more, visit: https://www.dol.gov/ofccp/regs/compliance/posters/pdf/eeopost.pdf. The position will be posted until a final candidate is selected or the requisition has a sufficient number of qualified applicants.