AI Engineer

xpu live; Dubai; On-site; AED 180,000 - 280,000; Full time

Job summary

xpu.live in Dubai, UAE is hiring an AI Engineer with strong Python skills to join our growing AI startup. You will build backend services and APIs with Python and FastAPI, design RAG pipelines, and integrate LLMs with databases and tools to deliver grounded responses.

The role requires hands-on experience with LLMs, embeddings, vector search, and tools like LangChain or LlamaIndex, plus knowledge of PyTorch and GPUs. On-site attendance in Dubai is expected, with remote options by arrangement.

Qualifications

• Proficient in Python with API development experience.

Responsibilities

• Develop backend services and APIs using Python and FastAPI
• Design and implement RAG pipelines for document retrieval and grounded response generation
• Integrate language models with databases, tools, and product services
• Evaluate and improve response quality, inference speed, and service reliability

Skills

Python programming

LLMs & embeddings

LangChain or LlamaIndex

HuggingFace & Transformers

PostgreSQL & Redis

Vector search tools

Git, Docker, Linux

GPU deployment (vLLM/llama.cpp)

PyTorch

Tools

FastAPI

PostgreSQL

Redis

Qdrant

pgvector

FAISS

LangChain

LlamaIndex

Docker

Linux

PyTorch

vLLM

llama.cpp

Git

Job description

At xpu.live, we are a growing AI startup developing AI-powered products and services. We are expanding our technical team and looking for an AI Engineer with strong Python skills and practical experience with large language models and RAG systems.

Responsibilities

• Develop backend services and APIs using Python and FastAPI
• Design and implement RAG pipelines for document retrieval and grounded response generation
• Integrate language models with databases, tools, and product services
• Evaluate and improve response quality, inference speed, and service reliability

Required Skills

• Strong proficiency in Python, with practical experience in software and API development
• Working knowledge of LLMs, embeddings, vector search, and RAG architectures
• Experience with at least one of LangChain or LlamaIndex
• Familiarity with Hugging Face and Transformers
• Experience with PostgreSQL and Redis
• Familiarity with vector search tools such as Qdrant, pgvector, or FAISS
• Ability to work with Git, Docker, and Linux

Experience deploying open-source models on GPUs using vLLM or llama.cpp, working with PyTorch, and optimizing inference speed and memory usage is a plus.

Location and Interviews

This position is based in Dubai, United Arab Emirates, and requires on-site attendance.

Interviews can be conducted in person or remotely, by prior arrangement.