Key skills
People ManagementAgentic ArchitecturesAPI DesignModel Context Protocol (MCP)OpenAI Function CallingSoftware Engineering FundamentalsTestingCI/CDDeveloper ToolsOrchestration for LLMsModern Data StackSQLAIOpenAIClaudeAgenticAnalyticsdbtCollaboration
About this role
dbt Labs is the pioneer of analytics engineering, helping data teams transform raw data into reliable, actionable insights. In this role, you will lead the AI Platform Team, focusing on building a scalable platform for AI agents and collaborating with product teams to ensure the platform meets specific agentic use cases.
Responsibilities:
- Build, lead, and coach a team of 5–8 engineers focused on building a robust, scalable platform for AI agents
- Architect the 'Agent-First' Experience: Move dbt beyond a UI-driven tool by building the APIs and services required for agents to reason, plan, and execute within the dbt ecosystem
- Define dbt’s MCP Strategy: Lead the development of dbt MCP tools that allow Claude, Codex, and other LLMs to fetch context, validate SQL, and understand metrics without leaving their development environment
- Bridge Platform and Product: Partner with product teams to ensure that the AI Platform provides the necessary primitives (memory, tool-calling, and reasoning loops) for dbt’s specific agentic use cases
- Coach engineers in building 'Agent-ready' codebases—focusing on deterministic outputs from a non-deterministic world and the nuances of tool-use optimization
- Drive Technical Excellence: Establish the standards for how agents should interact with dbt Cloud, ensuring security, governance, and auditability are never compromised for autonomy
Requirements:
- 3+ years in people management leading high-performing software engineering teams
- Experience with Agentic Architectures: You understand the lifecycle of an agentic loop (Plan -> Act -> Observe) and how to build infrastructure that supports it
- Technical Breadth in APIs & Protocols: Deep experience with API design, and ideally, familiarity with emerging standards like MCP (Model Context Protocol) or OpenAI Function Calling
- Software Engineering Fundamentals: You have a strong POV on how to maintain dbt's 'Analytics Engineering' rigors (testing, CI/CD) in an AI-driven world
- A passion for Developer Tools: You understand the workflow of a data engineer and how tools like Claude Code or Codex are changing that workflow
- Experience with Orchestration: You've built systems that manage state and context for LLMs
- Strategic Collaboration: You can partner with external AI labs and internal teams to ensure dbt is the preferred 'data context' layer for all major LLMs