About the Role
The company is looking for an AI Product Engineer (LLM & Agentic AI) to develop and deploy production-ready AI applications utilizing large language models (LLMs), retrieval-augmented generation (RAG), and Agentic AI. The successful candidate will contribute to the company's AI products and enterprise solutions, leveraging AI-powered tools to enhance operational automation, efficiency, and innovation.
Ready to apply for roles like this?
Unlock the company name and direct application link. Subscribers get instant access to fresh jobs across Dubai, Abu Dhabi and Riyadh, many with visa support.
Unlock employer & apply directly
Key Responsibilities
LLM Application Development
- Design and develop enterprise AI applications powered by proprietary and open-source LLMs.
- Build advanced RAG pipelines for ingestion, chunking, embeddings, indexing, retrieval, reranking, and grounded responses.
- Develop AI agents capable of tool calling, multi-step task execution, and integration with enterprise systems.
- Implement multi-agent and workflow-based AI solutions with reliable execution and error handling.
- Develop structured output pipelines, prompt templates, and context-management strategies.
- Build conversational AI capabilities supporting multilingual interactions, particularly Arabic and English.
LLM Platform & Inference Engineering
- Integrate multiple AI providers through OpenAI-compatible APIs.
- Deploy and serve open-source LLMs using inference frameworks such as vLLM.
- Work with cloud-hosted and on-premises inference infrastructure.
- Optimize inference latency, throughput, concurrency, token usage, and cost.
- Implement model routing, fallback mechanisms, request logging, usage tracking, and API integration.
- Support deployments on NVIDIA GPUs and other AI accelerators, including Qualcomm Cloud AI platforms.
Production AI Engineering
- Build reliable backend services using Python and frameworks such as FastAPI.
- Design RESTful APIs, asynchronous workers, streaming responses, and event-driven integrations.
- Work with SQL and vector databases such as PostgreSQL, pgvector, Qdrant, or Milvus.
- Integrate AI services into enterprise workflows, knowledge systems, and third-party applications.
- Containerize applications using Docker and support Kubernetes-based deployments.
- Implement observability, automated testing, CI/CD, and production monitoring.
AI Quality, Evaluation & Governance
- Develop evaluation pipelines measuring answer quality, grounding, hallucination rates, latency, and cost.
- Implement prompt-injection defenses, access controls, PII protection, and content guardrails.
- Support secure AI deployments aligned with enterprise privacy requirements.
- Evaluate and benchmark new LLMs, embeddings, reranking models, and inference approaches.
- Troubleshoot production issues across application, model, and inference layers.
Product Ownership & Collaboration
- Collaborate with product managers to convert requirements into functional AI capabilities.
- Own assigned features from technical design and implementation to deployment and monitoring.
- Deliver maintainable, reusable components rather than isolated POCs.
- Participate in architecture reviews, technical documentation, code reviews, and product demonstrations.
Requirements
Required Technical Skills/Experience
- 3–6 years of relevant experience.
- Strong programming skills in Python.
- Hands-on experience integrating and developing applications using LLM APIs or open-source models.
- Proven experience with RAG architectures and vector databases.
- Experience with at least one agent or LLM orchestration framework such as LangGraph, LangChain, or LlamaIndex.
- Strong API development experience using FastAPI or similar frameworks.
- Familiarity with Docker, Linux, Git, and production deployment workflows.
- Experience with SQL databases and asynchronous backend processing.
- Strong understanding of model parameters, tokenization, context windows, embeddings, and LLM evaluation.
- Ability to independently develop production-ready AI features.
- Fluency in both Arabic and English, with strong verbal and written communication skills.
Preferred Qualifications
- Experience with vLLM, Hugging Face Transformers, LiteLLM, or other inference frameworks.
- Experience building enterprise AI agents with tool calling, memory, and workflow orchestration.
- Knowledge of speech AI pipelines including STT, TTS, and real-time voice agents.
- Experience optimizing Arabic LLM applications.
- Familiarity with model quantization, GPU utilization, and inference performance tuning.
- Experience working with NVIDIA or Qualcomm AI acceleration infrastructure.
- Knowledge of Kubernetes, Redis, Kafka, or distributed application architectures.
- Experience with AI security, prompt-injection prevention, and PII filtering.
