Data Science Intern

Foxway is looking for a paid Data Science Intern - Price Prediction for Refurbished Devices

Every year millions of smartphones are replaced long before they stop working. Foxway gives them a second life - we are one of Europe’s largest refurbishers of mobile devices. Pricing those devices correctly is what makes the circular model work commercially: price too high and a perfectly good phone never finds a new owner, price too low and the business case for refurbishing disappears.

We are looking for a Data Science Intern to join our growing Business Intelligence & Data team in Tartu and take ownership of one concrete project: improving how Foxway predicts prices for refurbished mobile devices.

This is not a shadowing internship and not a tidy teaching dataset. You will work on a live model that supports real commercial decisions, using millions of transaction records, device condition data and market signals. You will sit next to experienced data scientists who will mentor you, and in the same office as the business leaders who actually use the output.

This is a great opportunity to find out how your skills hold up against a messy, genuinely open data problem.

Tasks

The project: extend and improve Foxway’s existing price prediction model for mobile devices.

What will you do:

• Explore transaction, device-condition and market data to understand what really drives the value of a used device

• Engineer and test new features, including data sources the current model does not yet use

• Train, tune and validate models in Python, and benchmark them honestly against the existing solution

• Error analysis - where the model fails and why, analysis across segments such as device model, storage, condition grade and market

• Present your findings and recommendations to both data scientists and business stakeholders in Tartu

• Document your work so that what you build can be taken forward into production

You will have a Data Scientist as your mentor and regular check-ins, but the project is yours to run. The internship can also be shaped to fit your university internship or thesis requirements - tell us what you need.

Expectations

What we need from you:

• Currently studying data science, statistics, computer science, mathematics or another quantitative field (final-year BSc, MSc or PhD)

• Proficiency in Python for data analysis and modelling (pandas/polars plus at least one ML library), and working knowledge of basic SQL

• Solid grasp of the machine learning fundamentals (regression models, model validation, overfitting, how to tell a real improvement from noise)

• Curiosity about the business problem and willingness to ask “why” from people outside the data team

• Able to work on site in Tartu (flexible work schedule)

• English as a working language; Estonian is a nice extra, not a requirement

• We do not expect you to have production ML experience

Nice to have

• Gradient boosting libraries (LightGBM, XGBoost), Git

• Interest/experience in pricing, e-commerce, marketplaces or sustainability

Practical details

• Internship period: The internship is expected to last for a minimum of 4 months. The exact start and end dates can be agreed upon based on your study programme and internship requirements.

• Paid internship

• Flexible work schedule from Mon-Fri, approx. 20 hours per week

• Starting time: preferred from October, flexible depending on the university schedule

• Location: on-site Tartu

We are reviewing applications on an ongoing basis, so it is worth applying early.

How does Foxway operate?

Foxway is a European technology company enabling a more circular way of using tech. Our mission is to make sustainable technology accessible to everyone: through Foxway Recommerce Mobiles we give smart devices a second life, reducing environmental impact and supporting a more circular future.

Headquartered in Sweden and founded in 2009, we operate in eight European countries with an international team of nearly 1,200 people - close to 500 of them in Estonia. Tartu is home to a large part of our mobile refurbishment operations, which means the data you work with and the people who use your results are in the same building.

You would be working with our Business Intelligence & Data which works on developing data driven tools to optimize and automate pricing tasks and other business needs.