Kritika Javali

• Working on a Municipal Solid Waste Management project funded by U. S. Department of Energy, in collaboration with NREL and IBM. The research aims at classifying wastes using Machine Learning models trained on annotated data. • Developed a Python web scraper using BeautifulSoup and GoogleSearch API to automate the extraction of waste characterization reports from official websites across 3140 US counties. Achieved 75% reduction in data collection time. • Devised a Natural Language Processing system using NLTK to extract and summarize key challenges in the waste management sector, which is expected to improve the decision-making efficiency for waste management policies by 20%. • Programmed statistical and machine learning imputation techniques to fill data gaps, reducing missing data from 35% to 0%. • Led the development of a waste generation forecasting model using Random Forest algorithm with an RMSE of 4% • Engineering an MLOps pipeline on AWS with EC2 instances and storage buckets to deploy a YOLOv8 model for waste classification. • Presented a research poster at the “Waste to Advanced Resources Matter” workshop held at NCSU. 2 manuscripts under preparation.

• Created code using YOLO7 to detect humans in youth sports videos. Saved the data in JSON format to train Actor-Context-Actor Relation Network for Spatio-temporal Action Localization, which are used to create professional highlights. • Led a team of 3 interns to create a model aimed at detecting blank videos, resulting in 50% savings in storage space and reframing time.

• Built 3 applications that received hotel booking, car rental, and flight ticketing data through Storm and ingested into Hadoop using Kafka, which was stored into 2 Hive tables for each application and queried. • Implemented changes to JSON parsers and messages in existing applications according to customer’s needs. • Performed QA testing of developed code to verify that the JSON messages were parsed correctly and data was being stored in designated Hive tables.

• Implemented ‘Integrated Testing of Multi-Channel Analyzer PCB’ and its documentation. • Gained knowledge and proficiency in diverse testing phases, such as Bare Board Testing, Visual Testing, and Integrated Testing, as well as various testing categories, including EMC and EMI testing, and Temperature and Humidity testing.

• Courses: Design and Analysis of Algorithms Software Engineering Database Management Systems Object-Oriented Design and Development Neural Networks and Deep Learning Foundations of Data Science Automated Learning and Data Analysis Computer Networks Human-Computer Interaction Advanced Topics in Machine Learning

Attained the first rank (among 180 students) in the ECE department in my first year, and was awarded a scholarship.