Agentic AI Engineer (United States)

Senior Agentic AI Engineer

Embedded Engineering Engagement

We are hiring Senior Agentic AI Engineers to join a high-impact engineering team building the next generation of AI-driven software testing and developer workflows.

This is not a traditional ML research or model-training role . We are looking for strong software and systems engineers who know how to build with AI agents: designing agentic workflows, integrating agents into existing engineering systems, managing context effectively, and making AI-driven automation reliable enough for real-world use.

The ideal engineer is hands-on, highly autonomous, comfortable navigating a large codebase, and able to take an ambiguous problem from concept through implementation.

What You'll Work On

You may work across several closely related areas, including:

• Agentic workflow integration — Integrating autonomous AI workflows into existing engineering and test infrastructure.
• AI-driven test automation — Improving systems that generate, execute, analyze, and maintain automated tests.
• Context management — Building systems that provide agents with the right code, history, tools, state, and information at the right time.
• Developer-agent tooling — Improving how AI agents interact with repositories, build systems, test infrastructure, and engineering workflows.
• Reliability and evaluation — Debugging agent behavior, identifying failure modes, and improving the consistency and usefulness of agent-generated results.

Responsibilities

• Design, build, and integrate production-oriented agentic AI workflows.
• Develop tooling that enables AI agents to interact effectively with large codebases and engineering systems.
• Improve automated testing and test-generation workflows using LLMs and coding agents.
• Design context-management and retrieval strategies for long-running or complex agent workflows.
• Debug failures across agents, tools, builds, tests, and surrounding infrastructure.
• Build integrations between agentic systems and existing developer/test platforms.
• Evaluate agent output and develop mechanisms to improve reliability and reduce failures or unnecessary human intervention.
• Work directly with senior technical stakeholders to turn loosely defined problems into working systems.
• Own projects end-to-end and communicate progress, risks, and blockers clearly.

What We're Looking For

Required

• 5+ years of professional software engineering experience, or equivalent demonstrated depth.
• Strong Python and/or comparable systems/backend programming experience.
• Strong Linux and software-debugging fundamentals.
• Experience building or integrating LLM-powered applications, coding agents, or agentic workflows.
• Experience with APIs, developer tooling, automation, test infrastructure, CI/CD, or large software systems.
• Ability to understand and modify unfamiliar codebases quickly.
• Strong engineering judgment around reliability, observability, testing, and maintainability.
• Excellent written and verbal communication.
• Demonstrated ability to independently drive ambiguous technical work.

Particularly Valuable Experience

• Claude Code, Claude SDK/API, or similar coding-agent platforms.
• Agent orchestration, tool use/function calling, MCP, skills, subagents, or multi-step agent workflows.
• Context engineering, context management, retrieval, memory, or prompt/tool architecture.
• Automated testing frameworks or AI-generated testing.
• Large-scale developer infrastructure or internal engineering platforms.
• Android, wearables, AR/VR, or other device-oriented development environments.
• CI/CD, build systems, containers, telemetry, or production debugging.
• Experience embedding AI capabilities into an established engineering ecosystem rather than building standalone demos.

Deep ML, computer vision, and model-training expertise are not required.

The Engineer Who Will Thrive Here

We are looking for someone who combines strong engineering fundamentals with an AI-native way of working .

You should be comfortable receiving a problem rather than a detailed implementation plan, investigating the surrounding system, collaborating with stakeholders, and driving toward a working solution.

You communicate proactively, surface blockers early, document important decisions, and maintain visibility while working independently. This is a high-autonomy environment where ownership matters as much as technical ability.

Example Problems

You might be asked to:

• Integrate an autonomous testing agent with an existing test platform.
• Improve an agent's ability to understand and navigate a large repository.
• Build context-management infrastructure for multi-step agent workflows.
• Automate portions of test creation, execution, debugging, or maintenance.
• Determine why an agent succeeds on some workflows but fails unpredictably on others.
• Design tooling that lets an agent safely interact with existing developer infrastructure.
• Reduce the amount of human intervention required to complete complex engineering workflows.

Engagement

Long-term consulting engagement working directly with engineers at a major technology company.

Remote , with substantial collaboration across engineering teams.