Senior AI Engineer

Charterhouse are partnered with a fast-growing technology company at the forefront of AI-optimised hardware and infrastructure. This business is building the infrastructure that powers the next generation of AI. As their team expands, they are looking to hire a Senior AI Engineers to join a highly technical and driven engineering team in Dubai.

About the Role

The Senior AI Engineer will take ownership of designing and building scalable AI systems and pipelines centred around LLMs, leveraging Python and modern GenAI frameworks. A key part of the role involves developing internal RAG-based systems and compelling demo applications that demonstrate the capabilities of the company's proprietary hardware, working at the intersection of cutting-edge software and advanced silicon. The position also requires close collaboration with highly technical engineering teams to profile and optimise AI workloads across diverse hardware platforms, including custom accelerator architectures.

On the technical delivery side, the Senior AI Engineer will be responsible for optimising prompts, embeddings, retrieval strategies, and overall model behaviour for production-grade performance, as well as fine-tuning models for domain-specific use cases using approaches such as LoRA, qLoRA, or full fine-tuning. The role involves building and maintaining custom pipelines for document ingestion, chunking, embedding generation, and retrieval, while defining robust workflows for model versioning, reproducibility, and experiment tracking.

About You

The ideal candidate will bring strong Python programming skills and hands-on experience with data analysis libraries including Pandas, NumPy, and SQL. A proven track record delivering retrieval-augmented generation (RAG) solutions is essential, alongside experience working with vector databases such as Milvus or Chroma, and evaluation frameworks such as Ragas or DeepEval. Experience deploying and monitoring large-scale AI systems using tools such as ClearML or Kubeflow is highly advantageous.

Exposure to CPU, GPU, or custom accelerator architectures, including NPUs, TPUs, or ASICs, is strongly preferred and will be a differentiating factor for this client.