Our client is looking for a hands-on Engineering Lead who owns their own AI Platform technical execution, architecture, code quality, and delivery across three AI abilities areas: Document AI, RAG, and Agentic AI Platform. Writes code, reviews code, and personally implements the most critical components. Must handle AI-specific non-functionals: model latency, token cost, retrieval quality, agent reliability, and evaluation discipline. Collaborates with AI Research Lead on model selection and retrieval quality.What You Own:AI Platform technical architecture: service decomposition, retrieval-pipeline design, agent-framework design, evaluation harness, security, deployment topology.Hands-on implementation of architecturally sensitive, performance-critical, and security-sensitive AI components.Architecture decision log: model adoption, vector-DB choice, agent-framework selection, evaluation methodology. Visible to CEO, CTO, AI Research Lead, GPM.PM→Engineering loop: demand measurable acceptance criteria for AI features ("90% precision on test corpus," not "should be accurate"). Refuse unmeasurable requirements.Code review and engineering quality bar — correctness, security, evaluation discipline, cost impact.Lead and mentor the Mid Full Stack Engineer. Grow the engineering practice.AI non-functional posture: latency baselines, token cost per request, evaluation harness for RAG quality and agent reliability.Security and data posture: what customer data goes to which LLM provider, PII handling, on-prem/hybrid for data-sensitive clients.Reliability: incident command for AI Sev1s (model degradation, hallucination spikes, agent loops, LLM provider outages).What You Bring:7+ years of software engineering, 3+ at senior/lead level on production software products.Strong backend foundation (Node.js, Python, .NET, or Java) with full-stack capability.Production AI/LLM experience: has shipped at least one production AI feature (RAG, extraction, classification, agent workflow) and lived with its operational reality. Veto if absent.Production experience across Node.js, React, Angular, MS SQL, MongoDB, PostgreSQL + LLM API integration + at least one vector DB.Application architecture ownership for an enterprise web application or AI platform.Database and retrieval design: relational vs document vs vector stores. Has built or operated a retrieval pipeline.API integration with LLM providers and enterprise systems. Understands model-provider rate limits and failover.Performance and cost troubleshooting in an AI system (latency, runaway token costs).Code review and real engineer mentoring (specific person, specific growth arc).AI-engineering daily usage (Copilot, Claude Code, Cursor). Tested live in Stage 3.Configurable product instinct — multi-customer shared codebase.Egyptian or Middle East enterprise context.Arabic–English bilingual at a professional level.Nice-to-Have:NLP, document understanding, information retrieval, or knowledge-graph engineering background.On-premises LLM deployment or hybrid AI architectures.AI evaluation and benchmarking: golden-set discipline, A/B testing of prompts and retrieval strategies.Agent frameworks at production scale (LangGraph, AutoGen, CrewAI, custom orchestrators).Security certification or formal training.