/
From Data Challenges to Real-World Models: An Interview with Sneha Ajay, Data Scientist at Odido
From Data Challenges to Real-World Models: An Interview with Sneha Ajay, Data Scientist at Odido
Search Blog

From Data Challenges to Real-World Models: An Interview with Sneha Ajay, Data Scientist at Odido
Apr 24, 2025
This interview with Sneha Ajay highlights how university projects, persistence, and practical curiosity helped her break into telecom data science at Odido - where she now turns raw data into real business impact.
Introduction
The telecom sector is evolving rapidly, and with it, the demand for skilled data scientists is growing faster than ever. From churn prediction to personalized services, telecom companies like Odido are increasingly relying on data to drive business strategy. To understand how young professionals are stepping into this space, we spoke with Sneha Ajay, a recent graduate of the BSc Data Science program at TU/e and now a Data Scientist at Odido Nederland. Sneha shares how university projects, persistence, and a curiosity for practical challenges helped her land her current, temporary role.
From University Projects to Industry Impact
Sneha’s journey into data science started during high school, where she discovered a passion for mathematics and programming. Unsure about which career path to follow, she attended university open days and discovered that data science was the right fit for her.
"Once I saw what the program was about, I knew it was exactly what I wanted to study."
She enrolled at the Eindhoven University of Technology and immediately found herself thriving in the project-heavy environment.
"Honestly, around 85% of what I use at work today, I already encountered during data challenges at university. The only thing that was missing was data collection - on campus, we always got clean, ready-to-use datasets. At Odido, data wrangling is a much bigger part of the process."
Entering the Industry: Tips for Data Science Interviews
Sneha received a take-home assignment rather than a live coding session to prepare her for her Data Science experience at Odido.
"They gave me a task to build a prediction model, which I could complete at home in my own time. I really appreciated this approach - it felt much closer to real-life work and gave me the space to focus and do it properly."
Her advice? Focus on solving the task properly, and don’t let stress take over. She encourages students to treat these assignments like real business projects.
Data Science in Action: Household Subscription Modeling
At Odido, Sneha has worked on churn and prediction models, including a particularly interesting one that focused on household subscription optimization.
"We wanted to predict who was likely to buy into a shared subscription model. The challenge was identifying the correct target group. So, we identified potential customers as those who are on the same bill for previous products."
She emphasizes that telecom companies move fast - they want explainable models that can be deployed quickly and understood by business teams. SHAP plots and descriptive statistics of important features, she notes, are becoming popular for model transparency.
Current Trends & Staying Up to Date in Telecom Data Science
When asked about current trends, Sneha pointed to the rise of localized chatbots and internal AI copilots.
"We’re seeing companies building department-specific GPT tools - things that help customer agents or product teams get answers fast. At Odido, we even have our own internal ChatGPT for data scientists."
She believes prediction models will remain a huge area in telecom, especially as businesses look to stay ahead of customer needs and competition.
But with such rapid change, Sneha emphasizes that staying up to date is just as important as knowing the trends.
"Subscribe to AI News. It gives you a weekly overview of the latest tools, models, and trends. It’s the easiest way to stay current in a fast-moving field."
She also cautions students to build foundational knowledge instead of relying too heavily on AI tools:
"ChatGPT can be helpful, but it’s no substitute for truly understanding how code works."
Advice for Aspiring Data Scientists
Sneha had clear, honest advice for students starting out in data science:
"You won’t get it all at once. It takes time to understand how projects actually work - so be patient and don’t be too hard on yourself."
Her top three tips:
Learn SQL and querying - it’s essential.
Get comfortable with coding, even if it seems daunting.
Stay informed about tools and trends - especially in this fast-moving field.
She also encourages attending conferences like PyData, working on personal projects, and not losing hope during tough job markets.
"The market can be tough, but don’t give up. Keep building, stay curious, and talk to people. That’s what makes a difference."
Sneha’s story is a great example of how academic projects can translate directly into career opportunities - and how a thoughtful, practical approach can help young professionals thrive in competitive industries like telecom.