Data Scientist

Join our dedicated team at BRINEL | IQANTO as a Data Scientis t specializing in Deep Learning and Large Language Models (LLMs), where you’ll drive innovation by fine-tuning and deploying state-of-the-art models. You'll be central to the development and integration of Retrieval-Augmented Generation (RAG) systems and work with AWS services to deliver scalable and efficient AI solutions.

This role demands hands-on expertise in model training, evaluation, and deployment— leveraging frameworks and platforms like PyTorch and Hugging Face. Your understanding of data lifecycle management and ability to work across ETL pipelines and CI/CD workflows will ensure robust and production-grade outcomes.

Responsibilities

• Fine-tune and deploy LLMs using frameworks like PyTorch and Hugging Face, with focus on RAG techniques.
• Design, train, evaluate, and optimize deep learning models.
• Prepare, evaluate, and refine datasets to ensure high model performance; identify data gaps and suggest improvements.
• Develop and manage ETL processes to support model training and production pipelines.
• Implement and maintain CI/CD workflows for model deployment and updates.
• Work with AWS services (SageMaker, Bedrock, Lambda) for model training and serving; use GCP as needed.
• Document technical processes clearly, including UML diagrams and architectural overviews.
• Collaborate closely with cross-functional teams to ensure solutions meet business and technical requirements.

Qualifications

• Bachelor’s degree in Computer Science, Data Science, or related field
• Minimum of 2 years’ experience in data science and machine learning projects.
• Proficient in Python (PyTorch) and AWS (SageMaker, Bedrock, Lambda)
• Experience with LLM fine-tuning and working with RAG architectures
• Strong background in ETL and CI/CD for ML
• Experience in fine-tunning, evaluating, and deploying models
• Skilled in data evaluation and dataset design
• Proficient in English with excellent communication and documentation skills (UML, technical writing).
• Nice to have: GCP experience, Linux/Bash, statistical knowledge