Build and evolve our AI-augmented testing pipeline: from product spec use cases through test cases, test data, unit tests, and end-to-end Playwright automation, all generated via Claude Code, plus large-scale codebase improvement and refactoring through AST-based tooling (Java/Spoon, TypeScript/OXC, Flutter Dart Analyzer). Leverage frontier models (Claude, Gemini, Kimi, GPT) as core engineering tools, evaluating new model releases for coding, reasoning, and agentic capabilities, and integrating the best-fit models into team workflows as they evolve.