Senior Agentic AI Engineer

About the role

About CodeRoad

CodeRoad provides end-to-end software development services, helping businesses scale with ideal infrastructure solutions. From staff augmentation to dedicated IT teams and general software engineering, our nearshore technology services empower businesses to thrive in an ever-evolving digital landscape

About the Role

As a Senior Agentic AI Developer, you will serve as the technical backbone of our AI engineering initiatives, designing, building, and deploying production-ready AI agents and agentic workflows. You will architect autonomous systems leveraging cutting-edge LLM frameworks, Retrieval-Augmented Generation (RAG), and Model Context Protocol (MCP) to seamlessly connect AI agents with complex enterprise infrastructures and data ecosystems.

This role is critical to transforming business requirements into scalable, secure, and highly reliable agentic solutions on AWS. You will anchor technical execution and architectural strategy, establishing robust evaluation, observability, and human-in-the-loop guardrails while guiding nearshore development teams through the entire lifecycle of production AI deployment.

Key Responsibilities

-

Architect & Build: Design, develop, and scale autonomous agentic workflows and multi-agent systems using LangGraph, CrewAI, or Agent Development Kit (ADK).

-

Anchor Cloud Infrastructure: Own the cloud deployment and security architecture of LLM-powered applications on AWS, using Amazon Bedrock and scalable microservices.

-

Integrate & Orchestrate: Connect AI agents to enterprise APIs, data sources, and deterministic tools via Model Context Protocol (MCP) and workflow platforms like n8n.

-

Optimize Knowledge Systems: Build high-performance RAG pipelines and vector database integrations using Pinecone or Weaviate to optimize contextual retrieval and reranking.

-

Evaluate & Secure: Implement end-to-end evaluation, guardrails, and security measures—including task-success metrics, hallucination checks, and protection against prompt injection.

-

Monitor & Mentor: Lead system observability using tools like LangSmith, Langtrace, or AgentOps while mentoring engineering PODs in Python and AI development best practices.

Requirements

-

Experience: 5+ years in professional software engineering, with 2–3+ years dedicated to LLM application engineering and agentic AI systems.

-

Tech Stack: Advanced Python, AWS (Amazon Bedrock), LangGraph / CrewAI, Vector DBs (Pinecone, Weaviate), RAG architecture, and MCP integration.

-

Observability & Eval Tools: Hands-on experience with LangSmith, Langtrace, or AgentOps, along with evaluation frameworks for LLM groundedness and accuracy.

-

Soft Skills: Strong ownership mindset, technical leadership, and the ability to mentor developers and articulate complex architecture to non-technical stakeholders.