PEAX
AI Transformation Work Force
Co-leading an AI Engineering Transformation initiative at PEAX — designing a standardized, compliant agent framework with clear guardrails, human-in-the-loop patterns, and measurable impact on code quality and delivery speed.
The Challenge
Teams at PEAX were using AI tools inconsistently and without shared rules. That created risks for data privacy, code quality and security. PEAX needed one approach that can be audited and traced. It had to make the teams more productive and meet the security and privacy standards of the Swiss B2B market.
The Approach
Co-led the design and rollout of a standardized AI agent framework covering skills, patterns, MCP toolsets, and guardrails. Human-in-the-loop review was established as a hard requirement — no autonomous agent actions without oversight. The framework enforces strict data privacy controls: no production data in agents, no external model training, and agent actions logged without sensitive content. Evaluated leading coding agents, defined architecture and configuration templates, and developed a structured coaching and rollout concept for cross-team adoption. Additionally leading the introduction of AI-driven end-to-end testing, leveraging AI to generate, maintain, and evolve Playwright test suites — significantly raising coverage and reducing manual test authoring effort. Continued contributing as a Senior Frontend Developer to the Angular platform in parallel.
Regression tests run in minutes. They used to take days.
Impact & Results
Established a foundation for safe, scalable AI-assisted development across engineering teams. The framework enables measurably faster feature delivery and higher code quality through standardized tooling — while maintaining full compliance with data privacy, security, and ethical requirements. AI-driven E2E testing is significantly raising test coverage and reducing manual authoring effort. Deliverables include guardrail and architecture specifications, agent evaluation reports, configuration templates, and a phased adoption playbook with coaching materials.
My Role
Senior Frontend Developer and Co-Lead of the AI Engineering Transformation Task Force. Responsible for co-designing the agent framework and guardrail specifications, evaluating coding agents, defining MCP toolsets and usage patterns, and driving structured adoption across engineering teams. Leading the introduction of AI-driven end-to-end testing with Playwright. Continuing to contribute to the Angular/Nx frontend as part of the broader product organisation.
Got a similar challenge?
Tell me about your challenge. I'll share how I'd approach it — and a realistic timeline.