Machine Learning Data Scientist

Vijayanagalakshmi Kagitha

India *******************@*****.*** 938******* in/vijayanagalakshmi-k-b53653295 github.com/Vijaya-nagalakshmi SUMMARY

Results-driven Data Science enthusiast with hands-on experience in Python, SQL, and Power BI. Built and deployed ML and deep learning models for anomaly detection, forecasting, and object detection using YOLOv8 and CNN. Eager to apply data-driven solutions to real-world challenges and grow in a dynamic team.

EXPERIENCE

Data Scientist Intern

AiSPRY August 2024 - May 2025

Developed and deployed machine learning models for solar power anomaly detection using Isolation Forest and KMeans, achieving a 15% improvement in prediction accuracy.

•

• Engineered forecasting models using ARIMA and SARIMA to predict future sales, contributing to demand planning for a retail client. Developed and implemented YOLOv8 and CNN-based models to detect hazardous substances in metal scrap and automate highway inspection and maintenance.

•

• Applied AI techniques to enhance accuracy and streamline safety assessments in industrial environments. PROJECT

Highway Inspection and Maintenance using AI

April 2025 - May 2025

Built a YOLOv8-based object detection model to identify road elements like markings, kerb paint, crash barriers, and speed breakers with over 95% mAP, automating highway inspection.

•

Analyzed and preprocessed 2,000+ images through EDA and applied augmentation techniques using Roboflow to improve model robustness and generalization.

•

• Reduced manual inspection time by 70% and improved detection accuracy through real-time, AI-powered analysis using Python and OpenCV. Detect and Count Hazardous Substances in Metal Scrap February 2025 - March 2025

Developed an AI-based object detection system using YOLOv8n to accurately detect and count hazardous substances in metal scrap, automating manual inspection.

•

• Collected, preprocessed, and augmented image data with Roboflow to enhance model robustness and improve detection accuracy. Reduced inspection time by 60% and enhanced workplace safety by ensuring compliance with environmental regulations through real-time detection using Python and Streamlit.

•

Optimizing Inventory and Demand for Apparel Retail Success December 2024 - January 2025

Cleaned and processed sales data using Python, Pandas, and SQL; performed exploratory data analysis with Matplotlib and Seaborn to identify sales patterns and handle missing values and outliers.

•

• Applied time series forecasting models including ARIMA, SARIMA, Moving Average, and AutoRegressive to generate accurate demand predictions. Improved forecast accuracy by over 20%, reduced inventory holding costs, minimized stockouts, and enhanced supply chain efficiency with a MAPE below 10%.

•

EDUCATION

Bachelor of Technology in ECE

IIIT RGUKT ONGOLE • Andhra Pradesh • 2024 • 8.5

SKILLS

Programming: Python, SQL

Machine Learning & AI: Machine Learning, Deep Learning, NLP, Computer Vision, TensorFlow, PyTorch, Neural Networks, Artificial Intelligence

(AI), Large Language Models (LLMs)

Data Science: Data Cleaning, Feature Engineering, Exploratory Data Analysis (EDA), Data Visualization BI & Analytics Tools: Power BI, Tableau, Power Query Analytics & Forecasting: Data Modeling, Time Series Forecasting Cloud Platforms: AWS

Databases: MySQL

Contact this candidate