Artificial Intelligence Engineer (Ref: 196989)

Role purposeBuild and harden the AI layer of a customer-facing agentic-AI solution (two AI agents) and migrate it from PoV to production on the client’s .NET-based architecture. Hands-on delivery role focused on LLM integration, agent orchestration, retrieval pipelines, Responsible AI, and production readiness — working alongside backend and DevOps engineers to deliver a complete, production-grade system.Key responsibilitiesDesign and implement AI services in .NET / C#, integrating LLM providers (Azure OpenAI / equivalent) into the application architecture.Build and orchestrate agentic AI workflows (multi-agent / tool-using systems), including prompt design, context handling, and tool/API invocation.Re-platform PoV AI services, prompts, and retrieval pipelines onto the client’s approved .NET architecture (migration execution).Implement RAG (Retrieval-Augmented Generation) pipelines: document ingestion, embeddings, vector search, grounding, and response citation.Replace PoV mock AI/data interactions with production-ready integrations, working with backend teams on Oracle Fusion, JD Edwards, and other enterprise APIs.Implement and harden Responsible AI guardrails (content filtering, prompt protection, output validation, safety policies) for public-facing use.Build prompt templates, evaluation datasets, and AI testing frameworks to ensure response quality, accuracy, and consistency.Optimise AI performance (latency, cost, caching strategies, retry patterns) and ensure scalability and load readiness for production traffic.Contribute to security and compliance, including handling of sensitive data, access control context, and enterprise governance standards.Support channel integration (web, WhatsApp where applicable) and ensure AI behaviour aligns with user journeys (registration, onboarding, country restrictions).Implement logging, monitoring, and observability for AI services (prompt/response tracing, failure analysis, usage metrics).Collaborate with .NET engineers, DevOps, and QA to deliver a fully integrated production solution and document for client handover.Required skills & experience5+ years building production software in .NET / C# / ASP.NET Core.Hands-on experience integrating LLM / GenAI solutions into applications (chatbots, copilots, agent-based systems).Strong experience with Azure OpenAI / OpenAI APIs (or equivalent LLM platforms).Experience with RAG architectures, vector databases, embeddings, and semantic search.Familiarity with agent frameworks (e.g. Semantic Kernel, LangChain, AutoGen, or similar), preferably in .NET environments.Solid understanding of prompt engineering, evaluation, and AI testing practices.Experience implementing Responsible AI / AI safety controls in production systems.Strong API integration experience (REST, OAuth2, enterprise systems).Experience working in enterprise / regulated environments and delivering production-grade systems.Git, testing practices, clean architecture; strong English communication.Nice to haveMicrosoft Certified: Azure AI Engineer Associate (or similar).Experience with Semantic Kernel / Microsoft AI stack in .NET.Prior integration with Oracle Fusion, JD Edwards, or similar ERP/HCM systems.WhatsApp Business API or multi-channel conversational interfaces.Experience in public-facing AI assistants or recruitment / onboarding use cases.Deliverables / success criteriaProduction-ready AI layer deployed on the client’s .NET architecture, including LLM integration, agent orchestration, and RAG pipelines; backend integrations aligned; Responsible AI guardrails implemented; performance, scalability and security targets met; and full documentation and handover to the client at end of engagement. No post-delivery support is included.