Join our team as an AI Engineer embedded within a  squad undergoing an Agentic SDLC transformation. Your role is to configure, customize, and optimize the agentic tools and systems that power the squad's delivery — ensuring that AI agents are properly integrated into the development workflow, output quality is consistently high across every SDLC stage, and the integration roadmap evolves in step with the program's maturity. You are not just a user of AI tools; you are the person who makes them work precisely for this context, this team, and this codebase.

Key Responsibilities:

- Configure and customize agentic tools — including GitHub Copilot (agent and plan modes) — to fit the squad's specific context, codebase, and delivery standards.
- Design and maintain integration roadmaps that define how AI capabilities plug into each stage of the SDLC, from requirements and technical solution through development, testing, code review, and deployment — without disrupting the client's existing, mature DevOps pipeline.
- Ensure output quality upstream and downstream: review, validate, and refine what AI agents produce at each stage, identifying deviations early and adjusting configurations, prompts, and guardrails to keep quality consistent and trustworthy.
- Build and refine prompt engineering assets — including specification prompts, code generation instructions, test generation templates, and documentation workflows — tailored to the squad's architecture and ways of working.
- Conduct tooling audits and support value stream mapping sessions, identifying bottlenecks in the SDLC where the right agentic capability can eliminate waiting time and reduce lead time.
- Define and implement guardrails and ethics and compliance standards for AI usage within the development workflow, ensuring responsible and governed adoption aligned with the client's policies.
- Pair daily with developers, QAs, and BAs to embed agentic practices into real work items, coaching team members on effective agent use and gradually transferring ownership as the squad gains autonomy.
- Develop and maintain technical documentation that captures agent configurations, integration patterns, and best practices, feeding both the squad's day-to-day operations and the final Agentic SDLC Playbook.
- Continuously monitor AI system performance — adoption rates, output quality, cycle time impact — and iterate on configurations to drive measurable improvements every two-week cycle.
- Support problem-solving at the intersection of AI tooling and software delivery, acting as the technical reference when agents produce unexpected results or when new SDLC stages need to be instrumented.

Requirements for this Position:

- Proven hands-on experience configuring, customizing, and operating AI agents and agentic development tools such as GitHub Copilot (agent mode), Claude Code, Cursor, or similar platforms in real software delivery environments.
- Strong prompt engineering skills applied to SDLC contexts, including the ability to write and refine prompts for specification, code generation, test creation, code review, and documentation use cases.
- Solid understanding of agentic AI concepts: agent orchestration, memory integration, MCP (Model Context Protocol), guardrails, and quality control of LLM outputs.
- Sufficient software engineering background to understand architecture decisions, read and evaluate generated code, and credibly engage with developers and architects on implementation quality.
- Experience with structured output validation and eval frameworks for LLMs, with the ability to design checks that ensure agent outputs meet defined quality and compliance standards.
- Familiarity with modern DevOps pipelines and CI/CD practices — this role plugs agents into existing corridors, not rebuilding them.
- Ability to define, track, and report metrics on agentic adoption and output quality (cycle time, rework rate, AI usage per work item, output acceptance rate).
- Strong communication skills in English, both technical and business-oriented, for daily collaboration with the squad and program governance.

Nice to have:

- Previous experience working in Supply Chain, e-commerce, or SAP ERP contexts, with familiarity with the domain's processes, integration patterns, and compliance constraints that directly shape how agents must be configured and validated.
- Experience developing or extending custom MCP servers to connect agents with external tools, enterprise data sources, or proprietary APIs.
- Knowledge of LLMOps practices, including observability, tracing, and cost monitoring for AI systems in production.
- Familiarity with Spec-Driven Development (SDD) and its application within an agentic workflow.
- Background in legacy system modernization or large enterprise transformation programs, with practical experience navigating the coexistence of legacy systems and new agentic toolchains.