Jhansi Lakshmi

As a Java Software Developer at Oracle Cerner Healthcare, I developed and optimized healthcare software solutions using Java, Spring Boot, and AWS EC2. Experienced in Docker, Jenkins (CI/CD pipelines), and MySQL for building scalable applications. Proficient in Hibernate for database management. I have worked with LSTM for image captioning and predictive modeling using Random Forest and Linear Regression. Skilled in Google Colab and Jupyter Notebook for machine learning development. Familiar with Eclipse, JUnit, Mockito, Eclipse Debugger, and Tomcat for development and testing. Strong background in Agile teams and ensuring high-quality code through unit testing and integration testing.

Image Captioning with Conditioned LSTM Generators 08/2023 - 12/2023

• Image Captioning with Conditioned LSTM Generators

• Developed and implemented a Visual Question Answering (VQA) system using deep learning and Deep Neural Network methodologies

• Preprocessed and trained on extensive datasets to achieve high accuracy metrics in image understanding tasks

• Integrated model architecture that combines visual and textual features with attention mechanisms for enhanced performance

• Demonstrated proficiency in training, evaluation, and inference processes to analyze images and generate meaningful responses

Data Scientist for Black Friday Sales Prediction 01/2021 - 05/2021

• Black Friday Sales Prediction using Machine Learning

• Collected and cleaned historical sales data, and performed exploratory data analysis (EDA) to identify trends, correlations, and key factors driving Black Friday sales

• Engineered new features, such as customer lifetime value and product popularity scores, to enhance the predictive model's accuracy

• Tested and trained various machine learning algorithms (e.g., Random Forest, Linear Regression) to select the most accurate model for forecasting sales, fine-tuning it using evaluation metrics like RMSE

• Presented predictive insights to stakeholders through visual dashboards, enabling businesses to optimize inventory, marketing strategies, and revenue during the Black Friday season

Advanced Parking System
Utilized OpenCV for edge detection and filtering to identify lane markings in real-time video feeds.
Implemented the Hough Transform algorithm to detect lane lines, integrating sensor data for enhanced accuracy.
Trained a machine learning model with TensorFlow/PyTorch to analyze vehicle position relative to detected lanes.
Conducted extensive testing and optimization using Python, C, high-resolution cameras, and ultrasonic sensors to ensure system reliability and performance in diverse driving conditions and driver assistance.

• Managed operations for university stadium hosting Women’s NBA matches for the Dallas Wings, ensuring smooth event execution.
• Coordinated with sponsors and vendors, facilitating partnerships and ensuring all event requirements were met.