Lead Full Stack & Agentic AI Engineer

Job Description: Lead Full Stack & Agentic AI Engineer

Location : Reston, VA / Washington D.C ( 3 days onsite in Reston, VA)

Duration : Long Term Contract.

Experience Required: 8+ years (2+ in GenAI / Agentic AI) 

Role Summary: A hands-on technical lead who builds agentic AI systems, not just talks about them. You  will own multi-agent application maintenance end to end, lead a small team, and turn AI ideas into systems that run reliably in production.

 Key Responsibilities

Agentic Solution Ownership: Lead design and hands-on build of agentic AI applications, from use-case discovery through PoC to production. Agent Architecture: Build single- and multi-agent systems in LangGraph/LangChain: orchestration patterns, state and memory, retries, and human-in-the-loop checkpoints. Tools & MCP: Build tool-calling agents and MCP servers that connect LLMs to enterprise systems, APIs, and data.

RAG & Context Engineering: Design retrieval pipelines (chunking, embeddings, hybrid search, reranking) and manage context for long-running agent workflows.

Evals, Observability & Guardrails: Set up agent evals (task/trajectory, LLM-as-judge), tracing (LangSmith, Langfuse, OpenTelemetry), cost/latency tracking, and guardrails for prompt injection and PII.

Full-Stack Development: Build UIs in Angular/TypeScript and backend services in Node.js and Python (FastAPI), including streaming agent responses.

Production Deployment: Deploy containerized agent services on Azure OpenAI, AWS Bedrock, or Vertex AI with CI/CD.

Technical Leadership: Lead and mentor developers (including distributed teams), run design and code reviews, and set engineering standards.

Stakeholder Consulting: Work onsite with business and technology stakeholders to shape use cases, run PoC workshops, and explain architecture trade-offs.

Required Skills: 8+ years in full-stack development, including 2+ years hands-on in GenAI with at least one agentic system in production. 2+ years as a tech lead, plus client-facing or onsite delivery experience. Strong TypeScript, Angular, and Node.js; working proficiency in Python. Hands-on LangGraph/LangChain experience. Candidates should be able to walk through an agent they built and deployed. Deep experience with LLM APIs (OpenAI, Anthropic Claude, Azure OpenAI, Gemini): tool calling, structured outputs, streaming. Experience with MCP, vector databases (pgvector, Azure AI Search, Pinecone), and RAG patterns. Practical experience with LLM evaluation and observability tooling. Solid grounding in REST/GraphQL APIs, PostgreSQL/MongoDB, Docker, CI/CD, and at least one major cloud platform. Awareness of LLM security risks (OWASP Top 10 for LLM Applications). Clear communicator who can explain agentic AI to non-technical stakeholders and work independently onsite.

 Preferred Skills:

Experience with CrewAI, AutoGen, Semantic Kernel, OpenAI Agents SDK, or Claude Agent SDK. Agentic AI applied to quality engineering or test automation. Familiarity with A2A protocols and LLMOps practices. Relevant certifications (Azure AI Engineer, AWS GenAI/ML) or published work and talks on agentic AI.