Master Thesis: Data-Driven Discovery of Fault Precursors in Gas Turbines (Machine Learning)

Location: Finspång, Sweden

Company

hirify.global develops energy technologies focused on reliable, sustainable power generation, gas turbine services, decarbonization, modernization, and digitalization.

What you will do

• Analyze historical gas turbine operational and performance data together with documented fault events.
• Identify and engineer relevant features from operational signals.
• Develop data-driven and machine-learning methods to detect fault precursor patterns.
• Investigate fault evolution and compare precursor consistency across different turbine units.
• Evaluate how early developing faults can be detected.
• Interpret machine-learning results from both data-driven and physical engineering perspectives.

Requirements

• Master of Science student in mechanical engineering, aerospace engineering, engineering physics, or a similar field.
• Good programming skills in Python.
• Experience with machine-learning frameworks such as TensorFlow or PyTorch.
• Familiarity with time-series analysis, data analytics, anomaly detection, or machine learning.
• Knowledge of software development practices, including object-oriented programming, modular code, version control, testing, and documentation.
• Ability to work independently, collaborate with engineers and domain experts, and connect machine-learning results with physical understanding.

Nice to have

• Mechanical engineering knowledge related to gas turbines, turbomachinery, or thermodynamics.

Culture & Benefits

• Collaboration between the Performance and RDC/ISA departments, combining gas turbine engineering with data analytics and machine learning.
• Technical guidance from specialists while independently exploring ideas and developing thesis solutions.
• Employment benefits including reduced working hours, advance vacation, and a health care allowance.
• Potential access to a flexible workplace arrangement.
• Inclusive international organization with employees in more than 90 countries.

Hiring process

• Applications are reviewed on an ongoing basis.
• Applications should be submitted by 22 October 2026; the role may be filled earlier.