Senior AI Engineer

Who you are
Metropolis is seeking a Senior AI Engineer to join our Applied AI organization, a team purpose-built to rebuild how the company operates from an AI-first perspective. You will be the builder at the heart of this transformation, designing and shipping AI-powered tools and automation pipelines that replace manual, repetitive work across finance, operations, revenue, and beyond. This is a hands-on engineering role with enormous scope: you will own systems end-to-end, from the LLM prompt layer through to production deployment, working side-by-side with Process Analysts who understand the business deeply. If you want to see your work directly change how a company functions — not someday, but now — this is the role.

What you'll do
Design, build, and ship AI-powered automation tools and internal applications that transform manual business processes into AI-first workflows
Architect and maintain LLM-based systems, including prompt pipelines, agentic workflows, tool-calling integrations, and retrieval-augmented generation (RAG) setups
Build and maintain data pipelines that connect source systems (Snowflake, internal APIs, SaaS tools) to AI workflows, ensuring data quality and reliability
Partner with Process Analysts to understand how business functions work today, translate requirements into engineering specifications, and iterate based on real user feedback
Own production reliability for the systems you build — monitoring, alerting, and iterating on AI outputs to maintain quality over time
Evaluate and integrate emerging AI tooling (models, frameworks, APIs) to ensure Metropolis is always leveraging the best available capabilities
Contribute to internal standards for AI development: prompt management, evaluation frameworks, model selection, and system architecture
What we're looking for
6+ years of software engineering experience, with 2+ years of hands-on experience designing, building, and deploying LLM-based or AI-powered applications in production
Strong Python skills — you write clean, maintainable code and are comfortable building APIs, data pipelines, and backend services
Practical experience with LLM APIs (OpenAI, Anthropic, or similar), prompt engineering, and agentic patterns (function calling, multi-step reasoning, tool use)
Familiarity with data engineering fundamentals: SQL, data transformation, working with cloud data warehouses (Snowflake preferred)
Experience with workflow orchestration tools (e.g., Airflow, Prefect, or similar) and cloud infrastructure (AWS preferred)
A strong instinct for product quality — you care about whether the tools you build are actually useful, not just technically functional
Ability to work autonomously in a fast-moving environment, manage competing priorities, and ship iteratively rather than waiting for perfection
While not required, these are a plus:
Experience with vector databases, embeddings, or semantic search is a plus
Our Stack
Languages + Frameworks: TypeScript, React, Scala (principally), Java (limited)
Datastores: MySQL, PostgreSQL, Snowflake
Cloud: AWS
Version control: Git & GitHub
AI Tooling: Copilot on GitHub and Claude Code
Observability: Datadog