W2 GenAI Backend Engineer

• • • This role requires the candidate working on W2 directly through Insight Global to support our client while on contract. • • • Not looking for a solution architect, need hands-on experience with RAG architecture, will be reporting into the chief architect of the group Required Skills & Experience -10+ years of experience in back end development or data engineering -5+ years of experience in ML infrastructure or cloud engineering -3+ years of experience building AI/LLM-based applications, ideally in enterprise environments -Strong hands-on experience with AI/data stack: Cognitive Search, Azure OpenAI, CosmosDB, Azure Functions, AKS, Dataverse. -Hands on experience creating strategic approaches for automated remediation of hallucinations -Experience working through production issues while releasing AI applications/products -Proficiency in Python and backend frameworks (FastAPI/Flask/Django). -Deep experience with RAG pipelines, multi-agent frameworks (LangChain, LangGraph, AutoGen, CrewAI), and Semantic Kernel. -Familiarity with vector stores (FAISS, Pinecone, Weaviate, Azure Cognitive Search). -Understanding of observability, telemetry, and evaluation frameworks. -Great communication, documentation and critical thinking skills Nice to Have Skills & Experience -Azure Cloud experience -Experience building internal GenAI chatbots -Experience with GraphRAG, entity-based retrieval, and knowledge graphs. -Exposure to continuous LLM evaluation (DeepEval, G-Eval, custom pipelines). -Experience in full-stack environments (React, Next.js, shadcn) beyond just API support. -Working knowledge of React/Next.js to support API integration with the UI. Job Description Insight Global's client in the data center industry is building an internal AI assistant, into a next-generation enterprise AI platform that combines RAG, GraphRAG, and agentic AI patterns with secure access to enterprise data. We are seeking a GenAI Backend Engineer who can design and build the backend foundation for this platform—using modern frameworks, such as Azure Functions, Semantic Kernel, LangChain, LangGraph, AutoGen, CrewAI, CosmosDB, and other data/AI services. This role will be responsible for developing the backend services that power LLM workflows, connecting to multiple enterprise data sources, validating data quality, and enabling evaluation pipelines. While the role is primarily backend-focused, familiarity with modern front-end frameworks (React, Next.js, shadcn) is important to ensure tight integration between APIs and user experiences. Additional responsibilities include: ADDITIONAL DAY-TO-DAY: Backend & Multi-Agent AI Services -Build and maintain backend services for Nexus using Azure Functions, Semantic Kernel, and LangChain/LangGraph/AutoGen/CrewAI. -Develop APIs to orchestrate LLM workflows, manage state, and integrate with chat front ends. -Implement multi-agent pipelines with planning, reasoning, and execution flows. -Optimize performance, latency, and token usage across pipelines to ensure reliability through robust error handling, retries, fallback strategies, and scalable, resilient service design RAG & Data Integration -Design and maintain RAG pipelines (chunking, embedding, vector indexing, hybrid retrieval), and develop automated tests to validate data ingestion, embeddings, and retrieval workflows to ensure accuracy, consistency, and reliability. -Extend retrieval capabilities with GraphRAG and entity-driven reasoning. -Connect to multiple data sources: CosmosDB, Dataverse, APIs, Azure Cognitive Search, SharePoint, file/document stores. -Build ingestion pipelines that validate, clean, and test data before adding it to the knowledge layer. -Create automated evaluation methods to test retrieval accuracy, hallucination rate, and dataset quality. Observability & Evaluation -Implement telemetry for latency, retrieval hit/miss, hallucination, cost, and user engagement. -Build feedback loops to improve models and prompts continuously. -Integrate logging and metrics into observability frameworks (OpenTelemetry, Prometheus, Grafana). Security & Governance -Enforce RBAC, compliance, and secure handling of enterprise data. -Implement data masking, sanitization, and lineage tracking. -Collaborate with governance teams to align with data stewardship and compliance frameworks. Collaboration & Cross-Stack Enablement -Work with React/Next.js engineers to define API contracts, payloads, and session flows. -Support front-end integration by providing reusable services and documentation. -Collaborate with data teams to source, validate, and integrate datasets. -Partner with product and UX teams to ensure backend design supports intuitive, secure user experiences.