Which side projects can help you find a data internship?

A data side project is more than a way to improve your technical skills. It is also a strong interview tool, an asset on your CV and an excellent way to network. Here is a selection of projects, from the most complex to the most accessible.

1. Create a complete prediction model (advanced)

Example

Build a property price prediction model using public data, such as INSEE or Kaggle data.

Technical objectives:

• Collect and clean data (Pandas, NumPy).
• Engineer features (create explanatory variables).
• Model the data (scikit-learn, XGBoost or even PyTorch).
• Create visualisations and a final dashboard (Streamlit, Dash).

Why it is powerful:

• You demonstrate your ability to handle an end-to-end data pipeline.
• You can deploy it online using Streamlit Cloud to show at interview.
• It is a concrete, distinctive project on a subject everyone understands.

2. Create an interactive dashboard (intermediate)

Example

Analyse sales for a fictional e-commerce business using a Power BI or Tableau dashboard. You can also use Python and Plotly Dash.

Technical objectives:

• Clean and structure data.
• Create key performance indicators (revenue, average basket value and conversion rate).
• Develop a user-friendly dashboard.

Why it is useful:

• It is a highly visual project to show in a portfolio.
• Recruiters can immediately see your ability to make data understandable.
• At interview, you can explain your visualisation choices and start a conversation.

3. Conduct a statistical study on a subject you enjoy (intermediate)

Example

Analyse Spotify trends, football players’ performances or the highest-rated films on IMDb.

Technical objectives:

• Collect data through public APIs.
• Explore and visualise data (matplotlib, seaborn).
• Extract relevant insights and tell a story with the numbers.

Why it is relevant:

• This kind of project shows your intellectual curiosity.
• You can discuss it enthusiastically at interview and connect with the recruiter.
• You can share it on LinkedIn and start discussions in your network.

4. Enter Kaggle competitions (intermediate to easy)

Example

Enter a “Titanic Survival Prediction” or “House Prices” competition.

Technical objectives:

• Learn to model quickly within a real framework.
• Read other participants’ notebooks to improve.
• Work on concrete data science problems.

Why it is valuable:

• It is recognised by recruiters in the field.
• You can add your Kaggle profile to your CV.
• It is also a way to network with other data enthusiasts.

5. Create a small site or app around a dataset (easy)

Example

A small “DataFacts” site displaying interesting statistics on a particular subject, such as economics, sport or the environment.

Suggested tools:

• Streamlit, Flask or Gradio for the front end.
• Python and Pandas to manipulate data.

Why it is useful:

• It is a quick project with a strong visual impact.
• It can start a portfolio or become a LinkedIn post.
• You show a product approach sought by recruiters.

The networking bonus

Side projects do more than demonstrate your skills. They also help you make connections:

• You can share your project on LinkedIn or GitHub,
• exchange ideas with other enthusiasts,
• and attract the attention of recruiters or managers who may contact you directly.

It is a good conversation topic at an interview or networking event. You are no longer just “candidate X”: you become the candidate who made that great Spotify data project.

🏁 Conclusion

A data side project is more than a technical exercise. It is a strategic investment in your future: you build skills, attract recruiters’ attention and create a distinctive profile in a competitive field.

Choose a subject you enjoy, create your first project and share your work. That extra step can make a difference in finding your ideal data internship.