Master Thesis Sales Time Series Foundation Models for Forecasting (f/m/x)

BMW Group; München; Hybrid; EUR 60.000 - 80.000; Teilzeit

Zusammenfassung

An innovative automotive company in Munich is offering a Master Thesis opportunity focused on implementing time series foundation models for forecasting. Candidates should be studying a relevant technical field and possess strong Python programming skills. This role includes responsibilities like developing forecasting methods and documenting results. Flexible work arrangements are available.

Leistungen

Mobile work

Student apartments (subject to availability)

Qualifikationen

• Interest in and initial experience with large language models and machine learning.
• Strong problem-solving mindset, attention to detail, and collaborative style.

Aufgaben

• Support implementation of time series foundation models for forecasting.
• Build strong baselines and ensure reproducible experiments.
• Engineer features, handle seasonality, missing data, and outliers.
• Apply methods to automotive demand forecasting use case.
• Document methods, code, and results for a high-quality thesis.

Kenntnisse

Large language models

Time series methods

Machine learning fundamentals

Python programming

Numerical libraries (NumPy, SciPy, PyTorch, TensorFlow)

Ausbildung

Studies in computer science, mathematics, industrial engineering or related field

Jobbeschreibung

URL of this page was copied to paste buffer

Master Thesis Sales Time Series Foundation Models for Forecasting (f/m/x)

THIS MUCH DOING ISN'T DOABLE AT JUST ANY COMPANY.

SHARE YOUR PASSION.

We believe in creating an environment where our interns really can learn by doing and where they are given their own areas of responsibility right from the start of their time with us. That’s why our experts will treat you as part of the team from day one, encourage you to bring your own ideas to the table – and give you the opportunity to really show what you can do.

Our team at the BMW Group is driving the digitalization of sales and supports the transition to fully data-driven decision-making and sales management. We encourage you to contribute your ideas and support you in gaining practical experience during your thesis.

What awaits you?

• You will support the implementation of time series foundation models (e.g., Chronos, Time-LLM, TimeGPT) for zero-shot forecasting, including probabilistic outputs.
• Furthermore, you help to build strong baselines (e.g., ARIMA, gradient boosted trees) and ensure fair, reproducible experiments.
• In addition, you will engineer temporal and exogenous features, handle seasonality, missing data and outliers.
• Moreover, you support the application of these methods to an automotive demand forecasting use case.
• You will also help to document methods, code and results, and contribute to a high-quality thesis and internal report.

Please note that your thesis must be supervised by a university on your part.

What should you bring along?

• Studies in computer science, mathematics, industrial engineering or a related technical field.
• Interest in and initial experience with large language models, time series methods and machine learning fundamentals.
• Advanced knowledge of Python programming and experience with numerical libraries such as NumPy, SciPy, PyTorch or TensorFlow.
• Strong problem-solving mindset, attention to detail and collaborative style in an interdisciplinary team.

Would you like to support our team in developing advanced time series forecasting models and gain practical experience? Then apply now!

What do we offer?

• Mobile work.
• Apartments for students (subject to availability & only at the Munich location).

Do you have questions? Then submit your inquiry easily via our contact form . Your inquiry will be answered by phone or email afterwards.

We at the BMW Group place great importance on equal treatment and equal opportunities. Our recruiting decisions are based on the personality, experiences, and skills of the applicants. More about this here .

Master Thesis Sales Time Series Foundation Models for Forecasting (f/m/x)