Jump to content

You are now visiting the Sandvik Group website in English. Would you like to switch to another language site with selected content translated?

Image analysis using deep learning in discrete manufacturing

Sandvik Coromant in Gimo, Sweden offers a thesis project within the area of image analysis and deep learning.


Sandvik Coromant in Gimo is a world leader in the manufacture of cemented carbide tools for turning, milling and drilling. Our customers are found in the metal processing, automotive and aircraft industries, and our aim is to meet the markets' more stringent demands for new precision products with great durability.

In Gimo, outside of Uppsala, lies Sandvik Coromant’s largest production site. This site is highly digitalized and automated, yet there are endless possibilities of further improvements. Thus, we are offering a master thesis project in image analysis and deep learning.

Background and scope of the project

At our site in Gimo we are at the forefront of Industry 4.0 and due to this we were in 2018 recognized by World Economic Forum as a Global Lighthouse. However, being a part of the Global Lighthouse Network does not mean that there are no possible improvements left. We are constantly searching for new ways of implementing the latest innovations in order to increase automation, productivity, and quality at our site.

One highly interesting area of research is image analysis using deep learning. We believe that this technology may help strengthen our digital thread in manufacturing and therefore we are looking for someone to both determine whether the technology is mature enough and help find accurate implementations.

Work description

The project will shift between a literature review where the applicant determines relevant technical tools and practical tests in implementing these tools in our manufacturing processes. The goal is to deliver useful tools whilst simultaneously broadening our analytics toolbox and creating a clearer understanding of the current possibilities and limitations of said tools.

The applicant is expected to work semi-independently in finding relevant tools and their possible implementations. It is crucial that the study is well anchored in existing processes as that is the main success factor when it comes to implementations.

Student background

We are primarily looking for someone studying engineering physics, computational science, IT, or computer science. Most important is an interest in manufacturing combined with experience of deep learning and image analysis. The applicant must be fluent in Swedish and English as well as being able to work independently. As the work at least partially will be carried out in Gimo the applicant is expected to have accommodation within commuting distance, such as Uppsala or Gävle.

At Sandvik, we believe that diversity of experience, perspective, and background leads to a better environment for our employees, our business, and our customers. Now, we’re looking forward to meeting you!


The thesis project lasts 20 weeks, starting in August 2023.

As this project aims to find possible implementations of image analysis in our production, the project is expected to be carried out at least part-time in Gimo. It will also be possible to work part-time from our office in Uppsala.

Application and contact person

For more information about the thesis or to apply, please contact:

Måns Bengtson, phone: +46 70 354 72 64, email: mans.bengtson_c@sandvik.com

The selection is done continuously. Please send us your application as soon as possible, but no later than June 16, 2023.

Read more about Sandvik Coromant at the website www.sandvik.coromant.com

We would like your consent

Sandvik and our vendors use cookies (and similar technologies) to collect and process personal data (such as device identifiers, IP addresses, and website interactions) for essential site functions, analyzing site performance, personalizing content, and delivering targeted ads. Some cookies are necessary and can’t be turned off, while others are used only if you consent. The consent-based cookies help us support Sandvik and individualize your website experience. You may accept or reject all such cookies by clicking the appropriate button below. You can also consent to cookies based on their purposes via the manage cookies link below. Visit our cookie privacy policy for more details on how we use cookies.