PARIDHI PARAJULI

ad4lmg@r.postjobfree.com linkedin.com/in/paridhi-parajuli/ Github Medium 256-***-**** SKILLS

Python, SQL, R, Pandas, Numpy, Scikit-learn, Tensorflow, OpenCV, Statsmodels, Keras, PySpark, Jupyter, Git, Sagemaker, EC2, S3, AWS Lambda, MySQL, Segment Anything Model (SAM), Generative AI, Computer Vision EXPERIENCE

Graduate Research Assistant — NASA IMPACT Aug 2022 - Present

• Curated benchmark dataset tailored for machine learning applications, integrating large and complex geospa- tial and environmental justice data. Employed a spectrum of statistical, image processing, and machine learning techniques also utilizing foundation models on both structured and unstructured data.

• Developed interactive visualizations & custom dashboards using Python, Javascript, and complementary tools, enabling enhanced data exploration and multi-touch interactions for large geospatial datasets. Associate Data Scientist — Extensodata Pvt Ltd April 2021 - June 2022

• Contributed to the architecture design, development & integration of Income Identifier module from raw trans- actions & Foneloan Decision Information System’s Score Engine,ensuring alignment with business objectives.

• Conducted in-depth data analysis resulting in a 23% expansion of the loan customer base through identification of meaningful data patterns, detection of data anomalies, and product-specific research. Demonstrated adept communication skills in collaborative efforts with clients.

• Developed ETLpipelines using Pentaho and integrated with the existing system using Airflow. AWS AI/ML Internship — Genese Cloud Academy Aug 2020 – Dec 2020

• Applied machine learning algorithms and executed projects on AWS cloud platform, utilizing EC2, S3, Sage- maker, Amazon Rekognition, and AWS Lambda.

PROJECTS

Drowsiness Detection and Alert System — Developed an award winning real-time computer vision system for detecting drowsiness in drivers. Utilized eye aspect ratio and head pose estimation techniques to identify signs of drowsiness through in-car cameras, providing timely alerts to ensure driver safety. Link Smart Document Scanner — Developed an efficient Python program for Smart Document Scanning, utilizing advanced computer vision techniques such as perspective transform, edge & contour detection. Link Big Data and Machine learning framework for Diabetes — Developed producer-consumer scripts for real-time diabetes data from API to Kafka, conducted comprehensive data preprocessing, crafted an 87% accurate diabetes classifier using PySpark, and implemented a robust lambda architecture with advanced big data tools. Link AI Based Pneumonia Detection — Built a tool to detect pneumonia through chest X-ray images using Tensor- Flow, employing Principal Component Analysis for dimensionality reduction and a neural network for training and prediction. Achieved an impressive accuracy of 94.02%, precision of 96.76% and recall of 95.04%. Link PCOS Prediction Dash — Created an interactive PCOS prediction dashboard using Streamlit, implementing diverse machine learning algorithms to train, visualize, and evaluate models. Link Handwritten Digits Generator — Implemented a generative adversarial network (GAN) to generate handwritten digit images using the MNIST dataset for training. EDUCATION

Master of Computer Science, University of Alabama in Huntsville, GPA: 3.85 Expected May 2024 Bachelor’s in Computer Engineering, Tribhuvan University, GPA: 3.9 Nov 2021 - April 2021 Relevant Coursework: Machine Learning, Introduction to Big Data Computing, Survey AI, Database Systems AWARDS & LEADERSHIP

Grand Prize Winner - AGU Michael H. Freilich Student Visualization Competition 2023 Grace Hopper Celebration (GHC) 2020, 2023 Student Scholar - AnitaB.org 2020, 2023 IBM Digital Badge & Certifications - Data Analysis with Python, Machine Learning with Python 2020 DeepLearning.AI Certifications - Convolutional Neural Networks, Neural Networks and Deep Learning 2020