Job Description & Details
This is a hands-on engineering role focused on moving beyond basic LLM wrappers and building systems that can actually reason, plan, and execute multi-step workflows. If you enjoy orchestrating complex agent ecosystems using tools like AWS Bedrock, LangGraph, and MCP rather than just tweaking prompts, this position gives you the architecture freedom to build production-grade autonomous systems.
What You'll Actually Be Doing
You'll spend your days designing and deploying intelligent agents that do real workâinteracting with enterprise APIs, calling external tools, managing long-term memory, and collaborating via multi-agent patterns. Expect to spend a lot of time wrestling with edge cases in agent reasoning loops, reducing latency, setting up reliable vector search or GraphRAG pipelines, and making sure your agent outputs have robust security guardrails before they hit production.
The Core Tech Stack
You need a rock-solid background in backend development or data engineering paired with modern AI frameworks. You'll be living inside cloud-native AI platforms like AWS Bedrock, Azure AI Foundry, or GCP Vertex AI, while relying heavily on orchestration tools such as LangChain, LangGraph, LlamaIndex, or CrewAI. Deep familiarity with vector databases, semantic search, and emerging protocols like Model Context Protocol (MCP) and Agent2Agent (A2A) communication is non-negotiable for this stack.
Interview Expectations
Expect the hiring team to present a system design scenario where an agent fails midway through a multi-step workflow due to an API timeout or bad tool execution, and they'll want to know how you handle state recovery and memory management. They'll also likely grill you on how you implement strict security guardrails when allowing LLMs to autonomously call enterprise tools and databases, so be ready to talk specifics on latency optimization and prompt injection defense.
Application Advice
Your resume needs to clearly separate your traditional backend/data engineering tenure from your specialized Agentic AI experience. Make sure keywords like LangGraph, MCP, vector databases, RAG, and multi-agent orchestration are front and center in your project bullet points. Don't just list frameworks; highlight how you measured agent performance, handled production failures, and scaled LLM applications.