AI Specialist (Arabic Speaker)

We are looking for an experienced Arabic speaking AI Specialist to independently design, build, and deploy production-grade AI and GenAI solutions across diverse enterprise engagements. With 5+ years of hands-on AI engineering experience, you will own the end-to-end development of LLM-powered applications, agentic systems, and RAG pipelines while mentoring junior team members and actively shaping the AI practice at Beinex.

This role goes beyond implementation: you will evaluate emerging AI technologies, contribute to solution architecture decisions, lead client-facing technical discussions, and drive innovation across the full AI engineering lifecycle.

LLM Solution Design & Development

• Architect and deliver production-ready LLM applications — conversational agents, intelligent assistants, and RAG pipelines with advanced retrieval and re-ranking.
• Build multi-agent systems using LangGraph, AutoGen, or CrewAI; apply prompt engineering techniques to optimise model performance.
• Integrate LLM solutions with enterprise systems via REST APIs and event-driven architectures.

Model Fine-Tuning & Optimisation

• Lead fine-tuning (LoRA, QLoRA, instruction tuning) and benchmark foundation models (OpenAI, Mistral, LLaMA, Gemini) for specific use cases.
• Optimise for latency, cost, and throughput; design evaluation frameworks to measure accuracy, hallucination, and safety.

Agentic AI & Workflow Automation

• Design agentic workflows with tool use, memory management, planning loops, and human-in-the-loop controls.
• Build AI-driven automation pipelines across structured and unstructured enterprise data sources.

Vector Databases & Knowledge Infrastructure

• Design and manage vector stores (Pinecone, Weaviate, Qdrant, FAISS, pgvector) with semantic and hybrid search strategies.
• Maintain knowledge bases powering enterprise AI applications, ensuring accuracy and freshness of indexed content.

Technical Leadership & Mentoring

• Guide and review work of associate-level engineers; contribute to reusable frameworks and internal engineering standards.
• Lead client workshops, technical discovery sessions, and PoC demonstrations; produce clear solution design documentation.

Trend Monitoring & Innovation

• Evaluate emerging LLMs, multimodal models, and local inference runtimes; prototype new tools and share findings with the team.
• Contribute to Beinex thought leadership through internal sessions, write-ups, or external blogs and talks.

• 5+ years of professional experience in AI engineering, machine learning, or applied NLP roles — with at least 2 years focused on LLMs and Generative AI.
• Proficiency in Arabic with strong speaking and writing skills
• Strong Python proficiency; fluent with libraries and frameworks including Hugging Face Transformers, LangChain, LangGraph, FastAPI, and PyTorch.
• Deep practical knowledge of LLM fine-tuning, prompt engineering, RAG pipeline design, and agentic system development.
• Hands-on experience with vector databases (Pinecone, Weaviate, Qdrant, FAISS, or pgvector) and embedding models.
• Proven ability to deploy and productionise AI solutions in cloud environments (AWS, Azure, or GCP) using containerisation (Docker, Kubernetes).
• Experience integrating AI solutions with enterprise platforms via APIs, webhooks, or event-driven architectures.
• Strong understanding of responsible AI principles including hallucination mitigation, output evaluation, content safety, and model governance.
• Ability to communicate complex technical concepts clearly to both technical peers and non-technical business stakeholders.
• Track record of independently owning AI projects from scoping through to production delivery.
• Preferred Certifications:
• AWS Certified Machine Learning – Specialty, or equivalent cloud AI/ML certification (Azure AI Engineer, GCP Professional ML Engineer).
• Deep Learning Specialization
• LLM Engineering or Generative AI certifications from recognised platforms (Hugging Face, DeepLearning.AI, Databricks).
• Certified Kubernetes Application Developer (CKAD) — advantageous for deployment-focused candidates.