Sr. AI Agent Engineer

Job Title: Senior AI Agent Engineer - SDLC Optimization

Location: Mclean, VA (5 days onsite)

Job Summary: We are seeking a forward-thinking AI SDLC Engineer to transform our development processes from linear automation to an agentic, multi-agent workflow. The ideal candidate will design, build, and deploy specialized AI subagents (e.g., using frameworks like LangChain, CrewAI, or Claude Code) that collaborate to create production-grade software. You will enable AI to handle specialized tasks architecture planning, frontend/backend development, debugging, and security reviews while maintaining context isolation.

Responsibilities:

• Subagent Architecture & Creation: Design and implement task-specific AI subagents (e.g., "Code Reviewer," "SQL Expert," "Frontend Builder") with dedicated system prompts, tool access, and context boundaries.
• Orchestration & Workflow Automation: Build orchestration logic to manage the lifecycle of subagents (initialization, delegation, execution, and cleanup) using agentic frameworks.
• AI-Driven SDLC Integration: Implement agentic workflows across the entire SDLC, from requirements gathering and architecture design to automated testing, code review, and deployment.
• Context & Memory Management: Optimize AI performance by isolating noisy, long-running tasks into subagents, ensuring the main agent remains efficient and context-aware.
• Tooling & Integration: Integrate AI agents with internal development tools, including Bitbucket, Jira, ServiceNow and CI/CD pipelines.
• Performance Evaluation: Monitor, evaluate, and refine agentic workflows to improve code quality, reduce toil, and accelerate release cycles.

Required Qualifications:

• Experience in Agentic AI: Proven experience designing and creating AI agents and subagents, specifically using frameworks such as LangChain, or custom agent harnesses (e.g., Claude Code).
• Software Development Background: Strong background in software engineering (Python/FastAPI preferred) with a deep understanding of CI/CD, Git, and cloud platforms (AWS).
• LLM Proficiency: Extensive experience working with Large Language Models (LLMs) and prompt engineering techniques for task-specific agent behavior.
• Systems Design: Ability to decompose complex software tasks into smaller, manageable, and parallelizable sub-tasks
• Experience with Claude Code and creating custom subagent markdown/YAML definitions.
• Knowledge of AI safety protocols and adversarial testing.
• Experience with vector databases for agent memory management.