AI Engineer

Role Overview

Key Responsibilities

• Design and build RAG pipelines: document ingestion, chunking, embeddings, hybrid search and re-ranking.
• Build and maintain knowledge bases and knowledge graphs (for example Neo4j or GraphRAG approaches).
• Fine-tune LLMs (LoRA/QLoRA, PEFT) on domain-specific data and evaluate them against base models.
• Develop agentic AI workflows, including tool calling, multi-step orchestration and memory.
• Set up LLM evaluation frameworks to measure accuracy, hallucination rate, latency and cost.
• Deploy and operate models in production (LLMOps/MLOps), including monitoring, versioning and CI/CD.
• Apply security, data privacy and responsible AI practices suited to regulated banking environments.
• Work with architects, developers and client stakeholders to turn business needs into AI solutions.

Mandatory Requirements

• 5-7 years in software, ML or AI engineering, with at least 2 years building LLM-based applications.
• Strong Python; hands-on with LangChain, LlamaIndex or similar frameworks.
• Proven experience building RAG systems in production.
• Hands-on with vector databases (Pinecone, Weaviate, Milvus, pgvector, Azure AI Search or similar).
• Experience with knowledge graphs and graph databases.
• Proven experience fine-tuning open-source LLMs (Llama, Mistral, Qwen or similar) using Hugging Face tooling.
• Experience with commercial LLM APIs (OpenAI/Azure OpenAI, Anthropic, Google Gemini).
• Experience building agentic AI and multi-agent solutions, including MCP (Model Context Protocol).
• On-premises and cloud experience on Azure, AWS or GCP, plus Docker and Kubernetes.
• Solid grasp of prompt engineering, embeddings and LLM evaluation methods.
• Arabic NLP or multilingual model experience.
• Experience in banking, fintech or other regulated industries.