Pulgam Lakshmi

Developed a deep learning model using ResNet and EfficientNet with 95–97% accuracy on the EuroSAT dataset. Built an end-to-end pipeline with 30% faster inference and map-based visualization to classify 10 land use types, aiding in urban sprawl and forest area detection for real-world geospatial monitoring.

Designed and implemented a lightweight search engine using TF-IDF vectorization and cosine similarity to retrieve relevant documents based on user queries. Developed a ranking algorithm to return the top-k most relevant documents, enabling real-time query processing and retrieval.

Collected and cleaned time-series and categorical data from public sources, performing exploratory data analysis (EDA) to identify trends, outliers, and correlations. Built and deployed a fully interactive dashboard using Streamlit/Dash/Tableau, enabling filtering by date, country, and metric.

Designed and simulated an astable multivibrator circuit using the IC 555 to generate a 38 kHz frequency. Built and tested the IR jammer circuit using breadboard and verified signal generation with an oscilloscope.

Grade: 3.76/4.00

Pursued a Master's degree focusing on advanced data science concepts and applications. Gained expertise in statistical methods, machine learning algorithms, and data visualization techniques. Completed various projects applying theoretical knowledge to real-world data challenges.

Grade: 3.67/4.00

Completed a Bachelor's degree in Electronics and Communication Engineering. Developed a strong foundation in electronic circuits, communication systems, and signal processing. Engaged in projects that enhanced practical engineering skills and problem-solving abilities.