Senior Applied AI Engineer (AI Platform)

Join our AI Platform team as a Senior Applied AI Engineer. You will be responsible for end-to-end product engineering, including architecture, evaluation, integration, and deployment. You will build and ship agents that automate fund-operations workflows, evaluate and improve agent performance, own the AI application architecture, and run it in production. You will also make it a platform, not a project, by consolidating document parsing and extraction into this team and setting patterns for other engineers to inherit.

- Concevoir et expédier des agents qui automatisent les flux de travail réels des opérations de fonds, en les intégrant à nos services, notre modèle de données et notre système d'autorisation. - Évaluer et améliorer les performances des agents, en construisant la couche d'évaluation : cas de test basés sur des documents réels, suites de régression en CI, examen humain lorsque la précision est non négociable. - Posséder l'architecture de l'application AI, y compris l'orchestration et la conception multi-agents, les contrats d'outils, l'ingénierie de la mémoire et du contexte.

- Production: you own what you ship. Reading traces, diagnosing failure modes, tuning cost and latency, handling provider rate limits and fallback without drama - Experience in fintech, private markets, or another regulated, document-heavy domain is a plus, as is working under EU data-residency constraints - Backend: You have experience with TypeScript and/or Python - Experience: 5+ years building production software, including at least one agent or LLM-powered capability you took end to end and still owned once it was live - Platform mindset: you build for other engineers as much as for end users, and the standards you set get adopted because people trust you, not because they are written down - Pragmatic: you start from the business outcome, choose the deterministic solution when it's the right one, and know when good enough is good enough - Evaluation: you know how to make a non-deterministic system measurable, from test cases built on real data to regression suites and human review, and you can tell the difference between a model that improved and a benchmark that got easier - Agents: real depth in orchestration and context engineering — tool contracts, memory, RAG, multi-agent design — with Mastra, ai-sdk, LangGraph or similar. You've integrated agents into a real product and its authorization model, not built prototypes for someone else to productionise