Vibha Prasanna

I am a student at Ramaiah Institute of Technology, aspiring to build a career as a Data Scientist. Currently developing my skills in data analysis, statistics, and machine learning, I enjoy learning how to apply data-driven approaches to solve real-world problems.

I have been exploring projects involving data preprocessing, visualization, and predictive modeling, which have helped me gain hands-on exposure to tools such as Python, SQL, Pandas, NumPy, Scikit-learn, and Power BI.
I am eager to further strengthen my expertise and contribute to data-driven decision-making, predictive analytics, and AI-powered solutions, while continuing to learn and grow as part of an innovative, collaborative team.

Data Science using Python from Edu-Versity with Wipro as the credential platform partner

• AI-Optimized Apartment Operations Portal

Problem Statement:

Traditional apartment management involves inefficient manual processes, poor tenant communication, and lacks intelligent decision support. Property managers waste time on repetitive tasks while tenants experience delayed responses and limited self-service options.

Solution:

Built an intelligent apartment management platform that automates workflows using LLMs:

• Automated Operations: LLMs handle maintenance requests, scheduling, and tenant communications
• Smart Tenant Experience: AI chatbots and personalized recommendations improve resident satisfaction
• Administrative Intelligence: AI-powered insights assist managers with data-driven decisions
• Unified Dashboard: Centralized interface combining tenant services and management tools

Tech stack:

Backend: FastAPI + PostgreSQL for high-performance API and data management

AI Layer: Large Language Models for workflow automation and intelligent recommendations

Frontend: Modern responsive UI for tenant and admin portals

Architecture: RESTful API design with AI-integrated service layer

• Tinnitus cause finder using AI

Problem Statement:

Tinnitus diagnosis is complex and often delayed due to its multifactorial nature and subtle correlations with lifestyle, health conditions, and environmental factors. Healthcare providers struggle to identify root causes, leading to generic treatments rather than personalized interventions, ultimately affecting patient quality of life.

Solution:

Developed an AI-driven analytics pipeline that identifies tinnitus patterns and underlying causes:

• Symptom Pattern Analysis: Machine learning algorithms analyze patient data to detect correlations between tinnitus characteristics and potential triggers
• Lifestyle & Health Correlation: AI uncovers hidden relationships between symptoms and factors like diet, stress, medical conditions, and environmental exposures
• Early Detection Support: Predictive models assist healthcare providers in identifying at-risk patients before symptoms worsen
• Personalized Interventions: Data-driven insights enable targeted treatment recommendations based on individual risk profiles

Tech Stack

Machine Learning: Python-based ML pipeline for pattern recognition and correlation analysis

Data Analytics: Statistical analysis tools for symptom mapping and lifestyle factor correlation

Healthcare Integration: APIs for patient data integration and clinical decision support

Visualization: Interactive dashboards for healthcare providers to interpret AI insights and patient patterns

• SkyShield: Weather-Driven Flight Reliability Analysis for Leh Airport

Problem Statement:

Leh Airport faces frequent weather-related flight cancellations due to unpredictable mountain weather conditions, causing significant passenger inconvenience and financial losses. Airlines lack accurate tools to predict weather-induced disruptions, leading to poor planning, last-minute cancellations, and reduced operational efficiency.

Solution:

Built a predictive analytics system that improves flight reliability through weather intelligence:

• Forecast vs Reality Analysis: Compares predicted weather conditions with actual data to identify patterns and improve accuracy
• Flight Cancellation Prediction: Machine learning models predict cancellation probability based on weather forecasts and historical data
• Risk Assessment Matrices: Visual risk scoring system helps airlines make informed decisions about flight operations
• Proactive Planning: Enables airlines to optimize scheduling, reduce passenger disruption, and minimize financial losses through early intervention

Tech stack:

Machine Learning: Python-based ML models for weather pattern analysis and cancellation prediction

Data Integration: Weather APIs and historical flight data processing pipelines

Analytics Engine: Statistical analysis tools for forecast accuracy assessment and risk calculation

Visualization: Interactive dashboards and risk matrices for airline operations teams

Data Sources: Real-time weather feeds, historical meteorological data, and flight operations records