Kaushika Potluri

Current Title: Data Scientist

Current Employer: Prudent Technologies and Consulting

What do you do?: Build Power BI dashboards integrated with Azure SQL Server, streamline business reporting & reducing manual effort. I develop and deploy machine learning models on Azure SQL-backed datasets, improving churn prediction accuracy by 22%. Automate data workflows using Python and Azure Data Factory, enhancing pipeline efficiency and model refresh rates.

How did you find your current position? Referral.

Favorite part of your job? Team collaboration and leading a team.

Are there any skills or lessons learned while in your program that you use often? On the technical side, the coursework emphasized Python, SQL, R, statistics, and machine learning, which prepared me to build reproducible analyses, write efficient queries, and validate results rather than relying on surface-level metrics. Many projects were based on real-world datasets, so I learned how to handle messy data, design meaningful KPIs, and question assumptions instead of taking outputs at face value.

Thinking critically about data is probably the biggest one. I learned not to take numbers at face value - always check data quality, understand where the data comes from, and validate results (often by cross-checking with SQL or alternative logic). That habit has been incredibly important in real-world projects. Another skill I use often is communicating insights clearly. We spent a lot of time presenting findings and justifying assumptions, which translates directly to explaining dashboards, KPIs, or trends to stakeholders who may not be technical.

How did your program/degree prepare you for this role and career? 

My program gave me a very practical foundation that directly maps to the kind of work I do in this role. Through my degree in Data Analytics and Computational Social Science, I was trained to work end-to-end with data—from sourcing and cleaning raw datasets to modeling, analysis, and communicating insights to non-technical stakeholders.

On the technical side, the coursework emphasized Python, SQL, R, statistics, and machine learning, which prepared me to build reproducible analyses, write efficient queries, and validate results rather than relying on surface-level metrics. Many projects were based on real-world datasets, so I learned how to handle messy data, design meaningful KPIs, and question assumptions instead of taking outputs at face value.

What really prepared me for the career long-term was the focus on problem framing and decision-making. I wasn’t just learning tools—I was trained to ask why a metric matters, how it ties to business or operational outcomes, and how to translate analysis into actions. That mindset is something I use daily, especially when building dashboards, validating logic against SQL, or explaining results to leadership.

Overall, the program bridged theory and practice, which made the transition into a professional data role much smoother and helped me contribute with confidence early on.

What advice would you give to current or future students? Work on developing your skillset everyday!