AI Analytics Engineer (AI & Analytics Platform)

• Build and maintain context infrastructure: Translate institutional business knowledge into structured formats — business glossaries, DBT model enrichment, semantic layer definitions in Omni Analytics — so that AI tools can answer questions accurately, not just confidently.
• Design and run evaluation frameworks: Develop predefined test cases, accuracy benchmarks, and validation workflows that measure whether AI-generated insights are trustworthy. Own the feedback loop between eval results and context improvements.
• Build and orchestrate AI agent systems: Help design, build, and iterate on the agent architectures that power our analytics tools — including prompt pipelines, tool orchestration, query routing logic, and guardrails that determine when AI should answer autonomously vs. escalate for human validation.
• Experiment and evaluate: Test prompt configurations, agent behaviors, and model outputs across different use cases — using eval results and accuracy metrics to drive continuous improvement.
• Develop internal AI tooling and workflows: Build tools and automations that improve DS&A's own efficiency — identifying opportunities where AI can accelerate the team's work and executing on them.
• Build automated insight generation systems: Design and develop AI-powered systems that proactively surface patterns, anomalies, and meaningful changes in business data — delivering the right insights to the right people without waiting to be asked. Think less "answer questions" and more "anticipate them."

• Technically curious and AI-forward: You're energized by LLMs, prompt engineering, and the evolving landscape of AI tooling. You've experimented with tools like Claude, ChatGPT, or Cursor — and you're eager to build systems around them, not just use them.
• A builder at heart: You have a bias toward making things. Whether it's a prototype, a pipeline, or a quick script to test an idea — you default to building rather than theorizing. You may not have deep software engineering experience, but you're comfortable picking up new technical skills and exploring unfamiliar domains, especially with AI tooling accelerating what's possible.
• Analytically grounded: You're SQL-proficient and have experience with modern data tools (dbt, Databricks, Snowflake, or similar). You have strong intuition for when data "looks wrong" and can validate query logic and troubleshoot issues independently.
• Not married to legacy tooling: You're more interested in what's emerging than what's established. You evaluate tools based on what they enable, not how long they've been around — and you're quick to adopt new approaches when they're better.
• A clear communicator and strong writer: Context engineering is fundamentally a writing discipline. You can translate complex business logic into precise, structured documentation that both humans and LLMs can interpret.
• Business-minded: You're genuinely curious about how the business works — how we sell, how customers use the product, what metrics matter and why. You ask "what decision does this support?" not just "is the SQL correct?"

Browse similar jobs