Applications of Data Science

Content

Objective

The aim of the course is to teach how to manage, with engineering discipline, the whole life cycle of machine learning models from development to production, and to introduce the construction of systems based on large language models.

Course Content

The course covers the engineering practices needed to take data science and machine learning models into production: version control, testing of code and data, feature engineering and selection, MLOps principles and experiment management, pipeline automation with continuous integration, model deployment through APIs and containers, model monitoring in production and continuous delivery. The final part is devoted to large language models: foundations, prompt engineering, advanced techniques, and LLM-based system architectures (LLMOps and agents).

Course Learning Outcomes

Ability to set up a reproducible workflow using version control and tests for code and data.
Ability to carry out feature engineering and experiment management.
Ability to automate data and model pipelines with continuous integration.
Ability to deploy models using APIs and containers, to monitor them in production and to practise continuous delivery.
Ability to understand the foundations of large language models and to design prompt engineering and LLM-based system architectures.

Teaching and Learning Methods

Lesson, case studies, data applications and exercises

References

Course page: https://gitlab.com/onayg/vm536
Foundations of Data Science: Avrim Blum, John Hopcroft, and Ravindran Kannan

Weekly Contents

Assessment System

ECTS