Agentic AI Engineer

Design Autonomous Agents: Build single- and multi-agent systems that break down complex goals into smaller steps, plan execution paths, and take action.

Workflow Orchestration: Develop multi-agent roles (such as planners, retrievers, executors, and validators) that collaborate across business processes.

Tool & API Integration: Connect AI agents to external software, REST APIs, databases, vector stores, and cloud services (e.g., pulling live data, sending messages, or querying CRMs).

Memory & RAG Pipelines: Implement short-term context windows and long-term memory or iterative Retrieval-Augmented Generation (RAG) loops.

Testing, Evaluation & Observability: Monitor agent performance, track reasoning traces, debug step-by-step loops, and apply human-in-the-loop safeguards.

Security & Governance: Enforce compliance, guardrails, and responsible AI practices to prevent erratic behavior or data leaks.

Programming Languages: Python (primary), Java, asynchronous programming, and clean software engineering fundamentals (OOP).

Agent Frameworks: LangGraph, CrewAI, AutoGen, LangChain, LlamaIndex, or Semantic Kernel.

LLMs & APIs: OpenAI (GPT models), Anthropic (Claude), open-source LLMs, and prompt/context engineering.

Data & Storage: Vector databases, relational databases, and RESTful/microservice backends (FastAPI, Flask)