Responsibilities
- Design end-to-end agent experiences across documentation, analytics & growth
- Understand B2B operator needs across documentation access, data analysis, and growth recommendations
- Define agent reasoning, orchestration flows, and product-triggered actions for the full operator journey (beyond simple text responses)
- Own the roadmap & deliver AI-powered product features
- Lead the roadmap for the Mirakl Agent across its sub-agents: Documentation Agent, Analytics Agent, and Recommendation/Growth Agent
- Write clear specs (prompts, retrieval logic, APIs, orchestration flows, UX) and drive execution with engineering, data engineering, core, and analytics teams
- Shape and improve the agentic architecture
- Define how the orchestrating agent coordinates sub-agents, manages hallucinations, and triggers real product actions
- Contribute to retrieval/ranking pipelines and integrate core workflows (auto-filled forms, dynamic charts, actionable growth recommendations, partial auto-execution)
- Build dashboards, monitoring & feedback systems
- Deliver operator-facing dashboards and configuration UI to visualize, tune, and monitor agent behavior
- Implement evaluation frameworks, quality monitoring, and feedback loops
- Drive adoption & cross-functional collaboration
- Partner with PMM, CS & SE to create demos, training, and onboarding for customers adopting agentic experiences
- Collaborate across Core, Analytics, Data Engineering, and sub-agent owner teams
- Support early customers and promote Mirakl Agent internally and externally
Requirements
- 5+ years of Product Management experience in AI products
- Strong understanding of LLMs, RAG, embeddings, fine-tuning, prompting, structured reasoning, and multi-agent orchestration
- Experience working with AI/ML and Data Science teams on model evaluation, tuning, hallucination reduction, and reliability monitoring
- Have already shipped at least one agentic project in production (even a simple one, e.g. a ticket support agent)
- Comfortable writing specs for AI logic, orchestration flows, ranking rules, APIs, data pipelines, and front-end configuration tools
- Able to navigate strategic uncertainty, pivot comfortably, and structure complex, greenfield projects from scratch
- Strong B2B mindset, with the ability to operate as a mini GM AI rather than a classic feature PM
- Analytical, experimentation-driven mindset
- Fluent in English and French
