Agentic AI Engineer

Overview

The Agentic AI Engineer designs and builds the cognitive architecture of AI agents that interact with enterprise ERP systems.

What you'll do

• Design reasoning frameworks that decompose complex ERP tasks into structured decision trees.
• Build agent personas and behavioral profiles for multiple Rimini Solution domains.
• Architect dynamic tool selection strategies for MCP tool invocation based on context and confidence.
• Design conversation state management for long-running multi-interaction agent sessions.
• Develop multi-agent coordination patterns including A2A communication and supervisor/worker hierarchies.
• Translate OPA policy definitions into agent-actionable autonomy decision boundaries.
• Define evaluation and testing frameworks for agent behavior, including scenario-based and adversarial testing.

What you'll need

• 4-6 years of software engineering experience with strong Python and/or Java proficiency.
• 3+ years working with AI/ML systems in production, including LLM-based applications.
• 1+ years designing or building agentic AI systems (autonomous tool-using agents, not just chatbots or simple RAG).
• Experience with agent frameworks including LangChain/LangGraph, Pydantic AI, CrewAI, AutoGen, Semantic Kernel, or equivalent.
• Experience with tool-calling/function-calling patterns in LLMs (MCP, OpenAI function calling, Anthropic tool use).
• Hands-on proficiency with Pydantic AI as the primary agent framework and fluency in it.
• Experience handling human-in-the-loop workflows for pausing and human judgment.

Details

• Location: Hyderabad.
• Role is described as based in India.