We are looking for an experienced AI/Backend Engineer (or small team) to help design and implement a foundational setup for an enterprise “Agentic AI” assistant.The goal is to enable a prompt-based interface where business users can ask questions and trigger actions across internal risk management and document systems through a governed API layer. This is NOT a chatbot UI project.This project focuses on backend architecture, AI orchestration, and secure enterprise integration. Target application includes IBM OpenPages SaaS (GRC) accessed via REST APIs. The first phase will implement a proof-of-concept that allows an AI agent to retrieve and update structured records using approved APIs via a secure middleware service. Scope of Work (Phase-1: Foundational Setup) Build a secure middleware API service (AI Gateway) Integrate an enterprise LLM (Azure OpenAI or equivalent hosted LLM) Implement tool/function calling (agent actions) Connect to an existing enterprise system via REST APIs (query + update operations) Implement structured response schema (business entities like risks, issues, controls, tasks) Implement logging, traceability, and error handling Implement role-aware request handling (no hardcoded credentials) Ensure deployment architecture supports IP allow-listing Provide clear documentation and deployment instructions No UI required (basic Postman/test interface sufficient) Expected Deliverables Backend AI orchestration service (production-style codebase) Tool schemas and function calling implementation API mapping layer (business model → system APIs) Sample prompts and evaluation tests Security design notes Deployment guide (cloud-ready) Architecture diagram Required Technical Skills Core Backend & APIs Python (FastAPI / Flask) OR Node.js backend development REST API integration and schema mapping OAuth / token-based authentication handling Secure error handling and structured logging AI / LLM OpenAI / Azure OpenAI APIs Function calling / tool calling agents Prompt engineering for structured outputs Retrieval-augmented workflows (optional but preferred) Cloud & Infrastructure Azure (preferred) or AWS/GCP deployment Static outbound IP / networking configuration Containerization (Docker) Environment configuration management Architecture Knowledge Designing middleware integration layers Converting complex APIs into business semantic APIs Handling long-running workflows Designing safe update operations (write vs read actions) Nice to Have Experience in enterprise systems integration Experience with agent frameworks (LangChain, LangGraph, Semantic Kernel, or similar) Knowledge of approval workflows / audit logging Experience deploying internal copilots or assistants Responsibilities Design the AI orchestration pattern (not just code implementation) Implement secure tool-based action execution Prevent unsafe or unintended write operations Provide extensible architecture for future integrations Document how new tools/actions can be added later Ensure solution is maintainable by internal developers Engagement TypeShort initial phase (2–4 weeks) with potential long-term extension for full platform buildout. When Applying, Please Include Relevant projects involving LLM integrations or AI agents Example of tool/function calling implementation (code or architecture) Cloud deployment experience Preferred tech stack and why Estimated timeline for Phase-1 We are specifically looking for someone who understands enterprise-grade AI integration, not only prompt engineering or UI chatbot development