AI Engineer, Product Analytics

Job Description Product Analytics is building self-service tools and operating AI agents that influence product development; agents that monitor product health, surface anomalies, analyze user behavior, and produce the insights product leaders rely on. We are seeking an AI Engineer to build those agents for our Tax, Audit & Accounting product portfolio. You will take pre-trained models, LLM APIs, and analyst prototypes and turn them into tooling the team and its stakeholders can depend on, working inside the shared repositories, standards, and evaluation tooling the AI Engineering Lead owns.

Within 12 months we expect agents owning whole analytics workstreams. This role builds a meaningful share of them. The role reports to Director, Product Analytics.

Key Responsibilities

As AI Engineer, Product Analytics, you will be responsible for:

• Build and Integrate AI Components: Integrate pre-trained models, LLM APIs, and tool-calling and MCP-style integrations into the team’s analytics products, with guidance from senior engineers on the more ambiguous problems.
• Harden Prototypes into Production Tooling: Take an experimental notebook or a prototype and refactor it into a production-ready component with the error handling, guardrails, telemetry, and logging the team can rely on.
• Build the Data Pipelines Underneath: Prepare, clean, and structure the datasets your agents reason over, and build and maintain the pipelines that feed them. Contribute to how your product’s data is modeled and exposed for LLM-based reasoning.
• Build Within Shared Standards: Work inside the shared repositories, reusable components, and context and data-access standards the AI Engineering Lead owns, and register what you ship so someone with no relationship to you can find it and use it.
• Evaluate Your Own Work: Apply the team’s evaluation standard and definition of done: validation sampling, source traceability, prompt and model version documentation, and testing suited to non-deterministic model behavior.
• Keep Your Agents Healthy in Production: Diagnose failure and quality drift in the agents you own, conduct root cause analysis on moderately complex issues, and update context, prompts, and logic as data or business rules change.
• Contribute to Proactive Intelligence Systems: Contribute to the systems that watch your product and portfolio’s health and surface what needs attention before anyone has to ask, including the natural language interfaces, semantic layers, and feedback loops that let those systems improve over time.
• Work With Product and Engineering: Interpret the business, functional, and technical requirements for your area, and help translate ambiguous business needs into agent designs alongside senior engineers.
• Grow Your Craft: Take part in planning, code reviews, and team ceremonies; document your components and give clear status updates. Track how AI tooling is changing and get effective with new frameworks quickly. Curiosity and momentum carry more weight here than any specific tool on your CV today.

Required Qualifications You are a fit for the role of AI Engineer, Product Analytics if your background includes:

Required Experience And Skills

• Bachelor’s degree in Computer Science, Data Science, or a related field.
• 3+ years in software engineering, data engineering, data science, or analytics, with hands-on exposure to building or contributing to LLM-based agents or AI-native systems.
• Demonstrated personal investment in AI: you experiment with new tools and build things on your own initiative; with work you can point to.
• Proficiency in Python and strong SQL, with a solid grounding in programming concepts, design patterns, SDLC principles, and unit testing.
• Exposure to multi-component AI systems: RAG pipelines, agents with multi-step reasoning, or tool-calling and MCP-style integrations. Exposure to AI-adjacent infrastructure: vector databases, embeddings, semantic search, and retrieval pipelines.
• Experience refactoring experimental or research code into production-ready components following established patterns.
• Familiarity with MLOps and LLMOps practices and the lifecycle of AI-powered software, including testing, evaluation, and monitoring.
• Clear written and verbal communication, including the ability to explain technical concepts to non-technical stakeholders and to ask clarifying questions early.

Preferred Qualifications

• Product sense: an instinc…