Role Overview

We are seeking a visionary Senior AI Engineer to spearhead our "Agentic AI" initiative. This role focuses on moving beyond static LLM prompts toward dynamic, autonomous agentic setups that can reason, use tools, and collaborate to solve complex business problems. You will design and implement sophisticated multi-agent architectures that automate end-to-end workfl ows, ranging from complex fi nancial document analysis to automated customer interactions and fraud detection.

Key Responsibilities

● Agentic Architecture Design: Lead the design and implementation of agentic frameworks (e.g., LangGraph, AutoGen, CrewAI) to create autonomous agents capable of multi-step reasoning and planning.

● Tool Integration & Function Calling: Develop and optimize "tools" for agents to interact with, including internal APIs, databases (SQL/NoSQL), and external web services.

● Multi-Agent Orchestration: Build systems where multiple specialized agents collaborate (e.g., a "Researcher" agent passing structured data to a "Reviewer" agent) to ensure high-fidelity outputs.

● Cognitive Architecture: Implement advanced RAG (Retrieval-Augmented Generation) patterns within agentic loops, ensuring agents have access to relevant, up-to-date context from our document repositories.

● Production Deployment & MLOps: Deploy agentic workflows as scalable microservices within AWS, ensuring robust error handling, observability, and cost-efficient execution.

● Continuous Optimization: Establish evaluation frameworks to measure agent performance, reliability, and "hallucination" rates in production environments.

Key Responsibilities

● Agentic Architecture Design: Lead the design and implementation of agentic frameworks (e.g., LangGraph, AutoGen, CrewAI) to create autonomous agents capable of multi-step reasoning and planning.

● Tool Integration & Function Calling: Develop and optimize "tools" for agents to interact with, including internal APIs, databases (SQL/NoSQL), and external web services.

● Multi-Agent Orchestration: Build systems where multiple specialized agents collaborate (e.g., a "Researcher" agent passing structured data to a "Reviewer" agent) to ensure high-fidelity outputs.

● Cognitive Architecture: Implement advanced RAG (Retrieval-Augmented Generation) patterns within agentic loops, ensuring agents have access to relevant, up-to-date context from our document repositories.

● Production Deployment & ML Ops: Deploy agentic workflows as scalable microservices within AWS, ensuring robust error handling, observability, and cost-efficient execution.

● Continuous Optimization: Establish evaluation frameworks to measure agent performance, reliability, and "hallucination" rates in production environments.