AI/ML Engineer

Virtusa; Dubai; On-site; AED 300,000 - 480,000; Full time

Job summary

Virtusa is seeking an AI/ML-focused Data Engineer to build intelligent data pipelines for unstructured content and integrate with modern ML ecosystems. The role emphasizes PySpark and Python, with a focus on document classification, cleansing, quality metrics, and working with LLMs, vector databases, and RAG frameworks.

You will bridge data engineering and machine learning to enable AI-first applications, collaborating with AI architects and platform teams to design end-to-end AI data readiness

Qualifications

• 5+ years of commercial data engineering experience.
• Strong PySpark and distributed data expertise.
• Experience with unstructured data, OCR/NLP, and ML/AI tools.

Responsibilities

• Build robust scalable pipelines for unstructured documents using PySpark and cleansing techniques.
• Develop data workflows feeding LLM-based pipelines with RAG and vector embeddings.

Skills

PySpark

Python

ML/AI libraries

Tools

Transformers

LangChain

Hugging Face

FAISS

Redis

Job description

We are looking for an AI/ML-focused Data Engineer who brings deep expertise in building intelligent data pipelines for unstructured content and is experienced in integrating with modern machine learning ecosystems. The ideal candidate will have hands-on experience in PySpark and Python with a strong focus on document classification cleansing quality metrics and the ability to work with LLMs vector databases and Retrieval-Augmented Generation (RAG) frameworks. Candidates will play a critical role in bridging data engineering and machine learning enabling the development of AI-first applications across

Responsibilities

• Build robust scalable data processing pipelines for unstructured documents (PDFs emails forms etc.) using PySpark and document cleansing classification and enrichment techniques to prepare high-quality data for AI/ML applications.
• Develop and integrate data workflows that feed into LLM-based pipelines and support vector-based retrieval using RAG vector embeddings document chunking and metadata tagging for semantic search and question-answering closely with AI architect AI/Data engineers and platform teams to design end-to-end AI data readiness pipeline quality and model integration strategies clearly to both technical and non-technical Agile methodologies and CI/CD best practices to deliver continuously evolving AI capabilities

Required Skills

• Overall 5 years of commercial experience with 2 years in relevant role Strong proficiency in PySpark and distributed data experience in core Python including ML/AI libraries (e.g. Transformers LangChain Hugging Face FAISS etc.).
• Proven expertise in processing unstructured data and document intelligence (OCR NLP classification tagging).
• Familiarity with vector databases (e.g. Redis) and embedding models for RAG of LLM lifecycle including fine-tuning inference and prompt working in agile environments collaborating with cross-functional communication skills with the ability to interface with both technical and business stakeholders.