SDLC AI Engineer

SDLC AI

Key Responsibilities

• AI Agent Engineering & Tuning: Architect and maintain advanced AI agents for coding and automation within GitHub workflows . This includes the active building, performance tuning, and refinement of agentic logic to ensure optimal performance.
• Evaluation & Quality Assurance: Implement robust evaluation frameworks to ensure that all AI-driven solutions meet established technical standards and best practices before deployment.
• Standardisation (Prompt Ops): Develop and govern prompt libraries , instruction sets, and organizational standards to ensure consistency and reliability in AI usage across all engineering teams.
• System Orchestration: Integrate AI solutions seamlessly with CI/CD pipelines and Model Context Protocol (MCP) servers to create automated, end-to-end development cycles.
• Strategic Collaboration: Partner with cross-functional engineering teams to embed AI-driven processes directly into core SDLC methodologies.

Technical Requirements

• Programming Proficiency: Advanced expertise in Python or TypeScript . Professional experience with Rust, Golang, or Java is highly desirable.
• Ecosystem Expertise: Deep knowledge of the GitHub stack, Claude Code or Codex , including hands-on experience with CI/CD integration and complex automation tools .
• Architectural Insight: A comprehensive understanding of AI frameworks and agent-based architectures , moving beyond simple prompts to multi-step agentic workflows.
• Engineering Standards: The ability to write clear, reusable prompts and Skills while maintaining the highest coding standards and documentation.
• · Note:
• · have the candidates operated in large scale enterprise/industrial SDLC (large enterprises with complex process/tools). Given they need to apply their genAI skills in the context of a transformation from old to new…leading/guiding/designing how old style meets new style of engineering. Often we see inexperienced developers hopping onto Claude to be hyper productive but have limited insight how to apply that in the complex enterprise landscape like LBG.
• · Feedback for few profiles:
• · the first guy xxxxx - very terse, difficult to get coherent responses, very narrow exposure to SDLC - talked about agile but not the process of delivery - not a leader/shaper, a mid to junior level developer with claude on cv.
• · The second guy xxxxxx- had good/relevant experience some years ago, was better conversatonally, but lacked the kind of sharp/concise and relevant responses to direct questions about Enterprise SLDC, considerations towards agentic, etc. No doubt he has a rich career but don't see him facing into the CTO/CDAO leaders to address a particular complex assertion and value proposition. On reading the CV's i did have high hopes for those 2 but unfortunately the interviews lacked substance.