Shraddha Pandey

Tempe, AZ • 623-***-**** • ad4o9e@r.postjobfree.com • linkedin.com/in/pandey-shraddha

EDUCATION

M.S. Data Science, Analytics and Engineering, Arizona State University, Tempe, AZ Aug 2023 - May 2025 Analyzing Big Data, Data Mining, Statistical Machine Learning, Data Processing at Scale, Data Visualization 3.89/4 GPA B.Tech. Electronics and Communication Engineering, Nirma University, Gujarat, India Jun 2017 - May 2021

(First Class with Distinction) 8.34/10 CGPA

TECHNICAL SKILLS

● Languages: Python, C, C++, SQL

● Tools & Technologies: PostgreSQL, MS SQL Server, MySQL, Power BI, SQL Server Integration Services (SSIS), ETL, DWH, Jupyter Notebooks, MS Suites- Excel, PowerPoint, Word

● Data Analysis & Libraries: NumPy, Pandas, Matplotlib, Seaborn, Statsmodels, Scikit Learn

● Certifications: Microsoft Certified: Azure Fundamentals (AZ-900), Microsoft Certified: Azure Data Fundamentals (DP-900), Complete Data Science Bootcamp (Udemy)

PROFESSIONAL EXPERIENCE

Arizona State University, Tempe, AZ: Graduate Services Assistant, Computer Science Department Aug 2023 - Present

● Assisted 300 freshman engineering students with team projects and labs.

● Designed lab and test questions for 5 cohorts.

Cognizant, Bengaluru, India: Programmer Analyst Jun 2021 – Jun 2023

● Designed and implemented robust databases, optimizing schemas, constraints, and views for better query performance, resulting in a 20% reduction in query execution time.

● Created SQL Server Integration Services (SSIS) packages with complex transformations and mappings, increasing data integration efficiency by 30% and reducing processing time by 25%.

● Collaborated with Engineering team and conducted data migration process by implementing automated data validation checks, resulting in a 50% reduction in data discrepancies and ensuring a smooth and error-free migration. PROJECTS

Empirical Evaluation of Image-to-Image Translation using various GANs, Class Project Aug 2023 - Dec 2023

● Orchestrated an empirical evaluation project on Generative Adversarial Networks (GANs) for image-to-image translation, leveraging UC Berkeley Lab's Maps dataset; achieved a significant 20% enhancement in image quality and accuracy, driving innovation in computer vision research.

● Spearheaded the exploration of Pix2Pix GAN, CGAN, CycleGAN, StarGAN, and BicycleGAN, showcasing expertise in diverse GAN architectures.

● Achieved superior results with Pix2Pix GAN, emphasizing its U-Net architecture and adversarial training for 92% accuracy in image translation tasks, with a notable low RMSE of 0.047.

● Demonstrated a strategic blend of theoretical depth and practical relevance, contributing to the evolving landscape of computer vision and GAN research.

Deepfake Detection Using Convolutional Neural Networks, Class Project Aug 2023 - Dec 2023

● Led a team project that developed an advanced deepfake detection system utilizing Convolutional Neural Networks (CNNs), reducing the risk of fraudulent content by 75% and safeguarding brand reputation.

● Conducted extensive evaluations on the Celeb-DF dataset to enhance the model's performance across various deepfake manipulation techniques.

● Explored CNN architectures, including VGG16, for effective feature extraction, achieving an outstanding accuracy of 94% on the test set.

● Oversaw fine-tuning strategies, optimizing the model for unique dataset features, resulting in a consistent 15% increase in real-world detection accuracy. Actively championed ethical technology development, emphasizing a nuanced balance between security, privacy, and individual rights in deepfake detection, aligning with a principled approach Improved Automatic Speaker Verification, Class Project Aug 2021 - Dec 2021

● Spearheaded the design and implementation of an accurate (95%) and robust automatic speaker verification (ASV) system.

● Developed a customized Deep Learning architecture, combining Convolutional Neural Networks (CNN) and a modified version of the ResNet architecture, achieving a 95% accuracy rate.

● Contributed to academic advancement by publishing a research paper detailing the project's methodologies and outcomes through Springer.

Contact this candidate