Staff AI Engineer

About the Staff AI Engineer Role at Datadog

Datadog’s MCP team is actively scaling its MCP Services, an interface enabling both external agents like Claude and Cursor, and internal Datadog AI Agents, to seamlessly interact with Datadog data. As a Staff AI Engineer, you will be instrumental in evolving this agent-friendly tool surface, thereby facilitating advanced agentic workflows across various data types including metrics, logs, and incidents.

This role involves pioneering advanced evaluation frameworks for agentic systems and contributing to the development of next-generation agent-tool interaction models. This domain is rapidly evolving, characterized by significant ambiguity and high impact, making it an ideal opportunity for an individual passionate about shaping Datadog’s future in applied AI.

At Datadog, we value our office culture, recognizing the importance of relationships, collaboration, and the creativity fostered within our physical spaces. We embrace a hybrid workplace model, ensuring our Datadogs achieve a work-life harmony that suits their needs.

What You’ll Do

• Lead initiatives to enhance Datadog’s public-facing MCP server, empowering intelligent agents to discover and interact with our diverse services.
• Design and implement specialized agentic tool surfaces for both evaluation and production use, catering to a wide array of AI agents.
• Construct and maintain sophisticated evaluation pipelines to measure agent performance across critical Datadog workflows, such as investigations, incident triage, and metric queries.
• Investigate and resolve intricate failure cases by meticulously analyzing tool output, refining query parsing capabilities, and improving agent feedback mechanisms.
• Collaborate effectively with both Applied AI teams and various internal departments to establish and align on shared standards for tool integration and data access.

Who You Are

• An experienced Staff-level engineer with a robust background in applied AI, agentic programming, or extensive experience with LLM-powered automation pipelines, LLM orchestration frameworks (e.g., LangChain, LangGraph, CrewAI), and agent orchestration/tool-use systems.
• Comfortable navigating high ambiguity and fast-paced environments, capable of autonomously defining and prioritizing project direction.
• Proficient in constructing evaluation frameworks for LLM agents or general AI systems, including expertise in metrics design and data instrumentation.
• Bonus: Possess experience with the MCP standard or have contributed to developing agent-compatible tooling surfaces, and familiarity with building and evaluating ReAct agentic loops.
• Demonstrate strong systems thinking, with the ability to reason effectively across multiple agents, tools, and diverse user scenarios.
• Passionate about advancing the frontiers of agent-augmented software and eager to influence the evolution of new interfaces.
• Proven ability to integrate AI coding tools into daily workflows, and skillfully validate, critique, and refine AI-generated output.
• Motivated to elevate software engineering best practices through AI and committed to developing innovative AI-enabled products.

Benefits and Growth at Datadog

Datadog offers a comprehensive benefits package designed to support our employees' well-being and professional growth.

• Generous and competitive benefits package.
• New hire stock equity (RSUs) and employee stock purchase plan.
• Continuous career development and clear pathing opportunities.
• Employee-focused, best-in-class onboarding experience.
• Internal mentor and cross-departmental buddy program.
• Friendly and inclusive workplace culture.

Note: Benefits and growth opportunities may vary based on country of employment and employment nature.

About Datadog

Datadog (NASDAQ: DDOG) is a global SaaS business known for its rare combination of growth and profitability. Our mission is to dismantle silos and simplify complexity in the cloud age by facilitating digital transformation, cloud migration, and comprehensive infrastructure monitoring for our customers' entire technology stacks. Engineered by engineers for engineers, Datadog serves organizations of all sizes across a broad spectrum of industries. We foster professional development, embrace diversity of thought, encourage innovation, and champion work excellence to empower continuous growth. Join us and become part of a collaborative, pragmatic, and thoughtful community where we tackle challenging problems, take calculated risks, and celebrate each other's successes.

Key skills/competency

• Applied AI
• Agentic Programming
• LLM Orchestration
• Evaluation Frameworks
• Tool-use Systems
• Data Instrumentation
• System Design
• Cloud Monitoring
• Software Engineering
• Machine Learning