Frontend Engineer - Agentic AI
Hawk
- Località
- Munich, BAVARIA, DE
- Orario
- FULL_TIME
- Pubblicato
- 14 ott 2026
Prima di candidarti, verifica sul sito dell’azienda che la posizione sia ancora aperta e leggi le condizioni complete.
Descrizione della posizione
L’interfaccia è in italiano. I titoli e le descrizioni delle aziende rimangono nella lingua originale di pubblicazione, che può essere l’inglese.
About Us
Hawk is the leading provider of AI-supported anti-money laundering and fraud detection technology. Banks and payment providers globally are using Hawk’s powerful combination of traditional rules and explainable AI to improve the effectiveness of their AML compliance and fraud prevention by identifying more crime while maximizing efficiency by reducing false positives. With our solution, we are playing a vital role in the global fight against Money Laundering, Fraud, or the financing of terrorism. We offer a culture of mutual trust, support and passion – while providing individuals with opportunities to grow professionally and make a difference in the world.
Your Mission
As a Frontend Engineer at Hawk, you will build and maintain the user interface of our agentic AI Investigation Agent. This is the product where compliance analysts configure AI workflows, watch autonomous investigations run, and review, question and act on what the agent concludes.
The frontend is a modern TypeScript monorepo made of a host application and a growing set of independently deployable microfrontends. These remotes are designed to be built once and used in two places: our new shadcn/ui-based Investigation Agent and our existing Material UI-based Case Manager. Your job is to understand this codebase deeply, keep it healthy, and ship new features with care, from the component level all the way to how it is built and deployed.
What You'll Work On:
Features: Build product features end to end in React and TypeScript, such as workflow graph editors, investigation timelines, data-heavy tables and dashboards that make agentic AI output understandable for analysts.
Microfrontends: Develop and maintain Module Federation remotes, keep remote/host contracts stable, and make sure features fit into both the Investigation Agent and the MUI-based Case Manager.
Design system and data layer: Contribute to our shared shadcn/ui + Tailwind component library (documented in Storybook), and consume our Python/FastAPI backends through OpenAPI-generated clients with TanStack Query and Router.
Build, test and deploy: Work in a pnpm + Turborepo monorepo with GitHub Actions CI, Vitest and Playwright tests, and deployments to AWS including per-PR preview environments.
Collaboration: Work closely with Product, UX, backend engineers and Data Science, and use AI coding tools such as Claude Code or Cursor as part of your everyday workflow.
Your Profile:
3–5 years building and maintaining production React applications, with strong TypeScript and modern React skills (hooks, composition, performance, complex state and forms).
Clean frontend architecture: you structure code around features and clear boundaries. You keep UI components presentational and reusable, put data fetching and business logic in hooks and typed API layers, and keep server state separate from local UI state. You prefer composition, small focused components and strict typing, and you treat cross-module contracts as an API.
Design-system experience: hands-on work with a component-library approach, ideally shadcn/ui, Radix and Tailwind. Familiarity with Material UI is a strong plus.
Microfrontends and delivery: Module Federation or similar runtime composition, shared dependencies and host/remote contracts, plus the build and deployment basics behind them: Vite, monorepos (pnpm workspaces, Turborepo or Nx), CI pipelines (GitHub Actions)
Comfort with ambiguity: you're happy working on problems that aren't fully defined yet. You ask the right questions, make sensible assumptions, build a first version and refine it with Product and UX, without waiting for a perfect spec.
AI tooling: daily, practical use of AI coding assistants (Claude Code, Cursor, Copilot, etc.).
Communication: clear communication in English, and the ability to explain technical decisions to both technical and non-technical colleagues.
Bonus:
Some Python: enough to read a FastAPI endpoint or Pydantic model, run the backend locally, trace a bug across the API boundary, or open a small backend PR. You don't need to be a backend engineer.
Infrastructure as code: exposure to Terraform or AWS CDK.
Quality tooling: Storybook-driven development, visual regression testing or accessibility testing.
Auth: Auth0/OIDC integration in single-page apps.
Domain interest: UX for AI products, or previous work in fintech, RegTech or compliance.