Python Data Scientist: AI Model Training & Evaluation

Hire Feed; España; Presencial; EUR 45.000 - 65.000; Jornada completa

Descripción de la vacante

Hire Feed is seeking a Data Scientist/Analyst to join the team on a full-time basis. The role focuses on developing and maintaining high-quality Python code to train and optimize AI models, while evaluating performance and analyzing results.

You will collaborate with researchers and annotators to implement reinforcement learning with human feedback, refine reward models, and lead supervised fine-tuning with task-specific datasets.

Formación

• Proficiency in Python for data analysis and model development.
• Experience with data science and machine learning techniques.
• Ability to use business sense and analytical abilities to extract insights from public databases.
• Strong skills in bug fixing and creating thorough technical documentation.
• Excellent communication skills to clearly articulate reasoning and logic in Jupyter notebooks or similar formats.

Responsabilidades

• Design, develop, and maintain efficient, high-quality code to train and optimize AI models.
• Conduct evaluations to benchmark model performance and analyze results for continuous improvement.
• Evaluate and rank AI model responses to user queries across diverse domains, ensuring alignment with predefined criteria.
• Develop comprehensive explanations and rationales for evaluations, showcasing excellent reasoning and technical expertise.
• Lead efforts in supervised fine-tuning, including creating and maintaining high-quality, task-specific datasets.

Conocimientos

Python

Data science

Analytics

Bug fixing

Communication

Herramientas

Jupyter notebook

Descripción del empleo

Hire Feed is seeking a Data Scientist/Analyst to join the team on a full-time basis. The role focuses on developing and maintaining high-quality Python code to train and optimize AI models, while evaluating performance and analyzing results.

You will collaborate with researchers and annotators to implement reinforcement learning with human feedback, refine reward models, and lead supervised fine-tuning with task-specific datasets.