Nivas Verelli

At the Ronnia Langston Foundation, I improved how donor data was tracked across global programs by replacing manual reporting processes with automation. I built ETL pipelines using Python and SQL to clean, transform, and standardize data from multiple sources. The data was structured into reporting tables and connected to interactive dashboards. This reduced reporting latency, improved data accuracy, and helped leadership identify key gaps in donor retention and funding distribution.

At the Centre for Development of Advanced Computing (CDAC), I analyzed large-scale operational data to understand why service requests were breaching SLA timelines. Using SQL on 50K+ records, I identified key bottlenecks such as approval delays and uneven workload distribution. I worked with stakeholders to redesign workflows, creating more efficient TO-BE processes. I also built Power BI dashboards to track KPIs like processing time and SLA compliance, helping reduce delays by around 15%.

This project focused on improving healthcare accessibility by helping users identify possible illnesses and connect with the right doctors. We built a system that predicts diseases based on user-input symptoms using a machine learning model trained on symptom–disease data. The system mapped predictions to relevant specialists and enabled appointment booking. I worked on data analysis, pattern identification, and integrating the prediction model into the system. The project was later published in an IEEE conference.