Agentic Automation Architect

As Chief Agentic Quality Architect, you will orchestrate the transition from traditional scripted testing to an AI-augmented quality ecosystem. Your primary goal is to leverage agentic AI tools to generate, execute, and maintain high-fidelity test suites that keep pace with a development environment where AI agents are writing a significant portion of the production code.
WHAT YOU'LL DO
Take ownership of a new product or suite of products as a hands-on quality leader. Conduct a comprehensive audit of the existing test estate across all layers — unit, integration, API, UI, sanity, and regression — evaluating depth, coverage, and reliability of each suite.
Evaluate the current QA tooling, frameworks, and automation maturity against company quality benchmarks and coverage expectations. Identify systemic gaps, test debt, and high-risk areas with no automated coverage.
Produce a Current State & Gap Coverage Report that maps existing tooling, highlights missing coverage domains, and recommends the adoption tooling needed to close the gaps — the baseline for every roadmap that follows.
Define missing test cases and user journeys. Prioritise end-to-end automation across critical business flows using agentic automation patterns, with a deliberate emphasis on left-heavy coverage — maximising unit and integration depth first, and extending rightward through black-box UI and regression testing.
Architect and own the phased quality engineering roadmap, structured as three delivery horizons:
Two-Week Plan: Complete the current-state assessment; identify the highest-risk coverage gaps; stand up quick-win automation on the most critical user journeys; evaluate and select the tooling to be adopted.
One-Month Plan: Core regression coverage established across primary domains; CI/CD quality gates operational and integrated into the development pipeline; agentic test generation producing its first validated suites.
Three-Month Plan: End-to-end agentic automation live and self-maintaining; company benchmark coverage achieved; AI guardrails enforced across all automated agent output; quality metrics visible to engineering leadership.
Prompt-Based Engineering: Utilise AI agents to automatically transform business requirements and manual test cases into executable Playwright or Cypress scripts.
Synthetic Test Creation: Implement tools that autonomously generate test data and edge-case scenarios that human testers might overlook.
Autonomous Maintenance: Deploy self-healing automation frameworks that use AI to detect UI changes and update test selectors without human intervention.
Agent Regression Strategy: Design and own a comprehensive regression suite specifically tuned to catch the non-deterministic “hallucinations” or logic errors common in AI-generated code.
Behavioral Locking: Implement characterisation testing patterns to “lock in” the expected behaviour of legacy systems during AI-assisted refactoring.
Autonomous Output Validation: Define quality gates to validate the outputs of autonomous agents, ensuring they meet functional, security, and performance boundaries.
WHAT WE'RE LOOKING FOR
Automation Frameworks: Expert-level proficiency in Playwright, Cypress, or Selenium.
AI Tooling: Hands-on experience using LLMs (Claude, GPT-4, etc.) and agentic frameworks to generate code or automate workflows.
CI/CD Mastery: Deep understanding of integrating quality gates into AWS-based pipelines or similar environments.
Architectural Mindset: Ability to design "Behavioral Snapshots" to safeguard critical business logic during rapid transformations.
10+ years in QA automation engineering, SDET, or test architecture roles

Work arrangement
No