Job Title

What we do:GMG is a global well-being company retailing, distributing and manufacturing a portfolio of leading international and home-grown brands across sport, everyday goods, health and beauty, properties and logistics sectors. Under the ownership and management of the Baker family for over 45 years, GMG is a valued partner of choice for the world's most successful and respected brands in the well-being sector. Working across the Middle East, North Africa, and Asia, GMG has introduced more than 120 brands across 12 countries. These include notable home-grown brands such as Sun & Sand Sports, Dropkick, Supercare Pharmacy, Farm Fresh, Klassic, and international brands like Nike, Columbia, Converse, Timberland, Vans, Mama Sita's, and McCain.What will you do:We are looking for an AI Engineer (GenAI & Agents) to build, productionize, and operate GenAI solutions that improve business workflows and customer experiences across a large omni-channel retail environment. You will work hands-on across LLM application engineering (RAG, tool/function calling, agent workflows), quality evaluation, safety/guardrails, and reliable deployment at scale. In Brief: - Build GenAI applications (RAG + tools/agents) from prototype to production. - Integrate LLM workflows with enterprise data, APIs, and internal systems. - Implement evaluation, guardrails, privacy/security controls, and observability. - Optimize for latency, cost, reliability, and continuous improvement in production. Responsibilities:Build & ship GenAI solutions: - Design and implement LLM-backed applications using patterns like RAG, tool/function calling, workflows, and agent-like orchestration where appropriate. - Develop APIs/services for GenAI capabilities (chat, copilots, summarization, classification, content generation, knowledge assistants). - Build reusable components (prompt templates, tool registries, orchestration patterns, guardrail modules) to accelerate delivery. Knowledge & retrieval engineering (RAG) - Build ingestion pipelines for knowledge sources (documents, KB articles, policies, FAQs, runbooks) with metadata and refresh cadence. - Implement retrieval strategies (vector + hybrid search, reranking, filtering, citations) and tune chunking/embedding approaches. - Enforce permission-aware retrieval (RBAC/ABAC) and ensure answers are grounded with references. Agents & integrations: - Implement agent workflows that can call tools safely (search, ticketing actions, calculations, data lookups) with strict controls. - Integrate with enterprise services via APIs (identity, data platforms, operational systems, knowledge repositories). - Design human-in-the-loop patterns for sensitive workflows and escalation paths. Quality, evaluation, and safety: - Create evaluation harnesses and regression tests (groundedness, relevance, factuality, refusal behavior, latency, cost). - Implement safety guardrails (PII handling, prompt injection defenses, policy constraints, output validation, moderation where needed). - Establish feedback loops: capture user signals, label errors, and continuously improve retrieval and prompts. Production engineering & LLMOps: - Deploy and operate services with proper CI/CD, monitoring, and incident response. - Implement caching, rate limits, fallbacks, and cost controls. - Maintain documentation, runbooks, and operational KPIs/SLOs for GenAI services. Technical Competencies: - 4 years of software engineering / data engineering / applied AI experience with proven production delivery. - 1–2 years building GenAI/LLM applications beyond demos (prototype → production). - Strong stakeholder collaboration and ability to work in ambiguous problem spaces. - Comfortable owning systems end-to-end: build, deploy, monitor, and improve. Technical(mandatory): - Strong Python engineering (APIs, testing, async, packaging, clean code). - RAG fundamentals: embeddings, chunking, retrieval tuning, reranking, grounding/citations. - LLM integration: structured outputs, tool/function calling, context management, prompt design. - Evaluation: test sets, automated evaluation + human review loops, regression testing. - Security/privacy basics: PII handling, permissioned retrieval, audit logging, injection mitigation. - Production: CI/CD, containers, logging/monitoring, performance and cost optimization. Technical(nice to have): - Experience with LangChain/LlamaIndex (or similar orchestration frameworks). - Experience with vector databases and hybrid search implementations at scale. - Cloud experience (AWS preferred) and data platforms (Databricks/Spark is a plus). - Experience with observability tooling and LLM cost management/FinOps. - Experience with responsible AI governance, red-teaming, or formal safety reviews. Qualification & Experience:- Graduation or Masters in Statistics, Mathematics, Computer Science or equivalent - 4 years of software engineering / data engineering / applied AI experience with proven production delivery. - 1–2 years building GenAI/LLM applications beyond demos (prototype → production).