Principal Developer, Agentic Engineering

Job Summary

The Principal Developer, Agentic Engineering leads and mentors a small team of agentic engineers responsible for rapidly producing enterprise-grade products using agentic development practices. This role incorporates agents, tools, and skills produced by Euna's central AI team, and is the technical authority on how those capabilities get put to work: prompt architecture, orchestration, and human-in-the-loop workflow design for real product delivery. Beyond leading their own team's output, this person builds the patterns, practices, and reusable processes for applying agentic tooling that other agentic teams across the engineering org adopt, turning what works on one team into a shared standard for all of them. This position is critical to scaling agentic engineering at Euna from isolated wins into a repeatable, org-wide capability.

Responsibilities

• Lead and mentor a small team of agentic engineers, setting technical direction on how to apply available agents, tools, and skills to the team's product delivery
• Own the quality and correctness of code and agent output across the team, regardless of whether it was hand-written or agent-produced
• Own the full arc from product ideation through production, rapidly turning early concepts and prototypes into working, deployable software
• Partner with Product early and continuously to shape, validate, and refine ideas before and during build, and with DevOps/CloudOps to ensure what's built is production-ready, reliable, and operable
• Iterate quickly across ideation, prototyping, and delivery cycles, adjusting direction fast based on feedback, usage, and results
• Design and codify reusable patterns, practices, and processes for using agentic tooling (workflow design, subagent delegation, task orchestration, skill composition) that other agentic teams across the org can adopt directly
• Drive rapid, reliable delivery of enterprise products using agentic engineering, balancing speed with agent reliability, context management, and failure recovery at a systemic level
• Partner closely with the central AI team to stay current on new agents, tools, and skills as they're released, and provide field feedback on what's working and what's missing
• Partner with other Principal and Staff Developers and engineering leadership to align agentic practices with Euna's broader technical strategy and standards
• Diagnose and resolve systemic agent failure patterns, tooling gaps, and workflow friction that span multiple teams, not just the team they lead directly, partnering with the central AI team as necessary
• Translate business needs into agent behaviors, prompts, and workflow designs, communicating clearly across engineering, product, and executive audiences
• Actively develop the agentic practice of engineers outside their direct team through pattern-setting, review, and org-wide mentorship

Skills and Experience

• 10+ years of software development experience, including 3+ years focused on AI-integrated or agentic applications
• 5+ years leading or mentoring a team of engineers, with direct people or technical leadership responsibility
• Expert-level understanding of applying agentic development in practice — prompt architecture, agent orchestration, and workflow design — with the ability to set patterns others follow, without requiring the ability to build agent frameworks or MCP infrastructure from scratch
• Deep understanding of product management practices, with demonstrated ability to translate product ideation and prototypes into working, production-ready software rapidly
• Proven ability to iterate quickly across ideation, prototyping, and production cycles, comfortable operating end-to-end rather than handing off between stages
• Skilled at partnering directly with Product on the front end and DevOps/CloudOps on the back end throughout the software lifecycle
• 3+ years working with Claude-powered agents or similar LLM systems in a production environment
• Deep experience using the Claude Agent SDK or comparable agent frameworks (OpenAI Agents SDK, LangGraph) for multi-agent coordination, subagent delegation, and task management, at a scale spanning multiple teams or services
• Strong working knowledge of MCP servers, connectors, and skills as consumed capabilities — able to evaluate, compose, and get the most out of what the AI team builds, and to give informed feedback on what's needed next
• Proven track record designing autonomous workflows and human-in-the-loop patterns that hold up at enterprise scale
• Excellent problem-solving skills applied to agent reliability, context management, and failure recovery, including diagnosing and resolving systemic (cross-team) failure patterns
• Excellent executive communication skills; ability to translate complex agentic and technical considerations into business impact for non-technical audiences
• Bachelor's or Master's degree in Computer Science or related field, or equivalent industry experience demonstrating equivalent expertise

Nice-to-Haves

• Experience standing up a team-level agentic practice using an existing internal AI/tooling platform, rather than from scratch
• Contributions to internal pattern libraries, playbooks, or reusable agentic assets
• Prior experience working in or consulting with government/public sector technology environments
• Experience adapting CI/CD or quality gates for agent-produced code within an existing platform
• Familiarity with change management for teams transitioning from traditional to agentic development workflows