Lead AI Engineer

Lead AI Engineer (Multi-Agent Orchestration)—Remote---H1B,H4EAD,USC
strong MCP expert
We are seeking a hands-on Lead AI Engineer to spearhead the development of our AI Orchestration layer. This is not a high-level strategy role; we need a builder who is comfortable designing and implementing complex Multi-Agent systems where independent AI agents collaborate to solve multi-step tasks. You will be responsible for the "brain" of our platform, ensuring seamless communication, task delegation, and state management between various AI models and tools.

Core Responsibilities
Architect & Develop: Build robust AI Orchestration frameworks that manage multi-agent handoffs, parallel processing, and autonomous decision-making.
Agentic Workflows: Design and implement agentic patterns (e.g., Reflection, Tool-use, Planning, and Multi-agent collaboration) to solve non-linear problems.
Act as the subject matter expert to establish our internal AI Orchestration capabilities from the ground up.
Refine agent communication protocols to reduce latency and improve the accuracy of complex, multi-step AI outputs.

Technical Requirements
AI Orchestration Frameworks: Proven experience with LangGraph, AutoGPT, CrewAI, or Semantic Kernel.
Programming: Mastery of Python and experience with asynchronous programming (critical for multi-agent workflows).
LLM Integration: Deep understanding of LLM APIs (OpenAI, Anthropic, Gemini) and how to manage context windows and prompt chaining for agents.
Infrastructure: Familiarity with Vector Databases (Pinecone, Milvus) and CI/CD pipelines for AI-driven applications.
Core Fundamentals: Strong background in Java, OOPS, and scalable software architecture.

Preferred Experience:
Experience building autonomous agents that can perform real-world "tool-use" (executing code, calling APIs, browsing the web).
Knowledge of eval frameworks for measuring agent performance and reliability.