Ashish Sasi

Developed an AWS-hosted web application with an integrated ML backend for environmental data analysis.

Built data processing pipelines for food impact analysis and CO₂ emission calculations.

Created the frontend with JavaScript, HTML, and CSS, using D3.js to deliver interactive visualizations and dashboards.

Integrated engaging elements such as a Scratch-embedded game to enhance user interaction.

Implemented database querying to support nutrition-related calculations and insights.

Implemented Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for sentiment analysis on 650,000+ restaurant reviews.

Performed text preprocessing, filtering, and data cleaning to prepare large-scale unstructured review datasets.

Built and optimized models for label prediction on 250,000 records, improving classification accuracy and reducing noise.

Leveraged word embeddings and sequence modeling to capture contextual meaning and sentiment nuances in customer feedback.

Evaluated model performance with metrics such as accuracy, precision, and F1-score to ensure reliable results.

Built an Excel-based business intelligence model for sales forecasting and inventory optimization in a retail chain.

Applied time series forecasting techniques to predict product demand and improve stock planning.

Utilized Excel Solver for optimization, balancing stock availability with minimal holding costs.

Achieved a 25% reduction in stockouts while lowering excess inventory and improving supply chain efficiency.

Delivered actionable insights through BI dashboards to support management decision-making.