Sai Pratyush Varma Pinnamaraju

Results-driven Data Analyst with extensive expertise in AI-powered platforms and machine learning applications. Proven track record of leveraging technologies such as OpenAI GPT-3.5 to extract actionable insights and create scalable data solutions. Recognized for strong problem-solving abilities and effective collaboration, with a commitment to advancing data science through innovative AI/ML development.

• Data structures in python
• Programming for Everybody using Python
• Complete Guide to Power BI for Data Analysts by Microsoft Press

Stock Trading (July 2024– Present)  

• Utilize Pandas for data manipulation and pre-processing to prepare stock data for model training  effectively.  

• Implement TensorFlow to build and train LSTM neural networks for accurate stock price predictions.  

• Visualize results using Matplotlib to compare predicted stock prices with actual values for evaluation  insights.  

Created Sales Dashboard Using Power BI (July 2024-July 2024)  

• Completed sales data analysis project in Power BI, creating an interactive dashboard with engaging  visualizations.  

• Utilized DAX for advanced calculations, enhancing the depth of insights and metrics in the analysis.  

• Applied data modeling techniques to structure data effectively, optimizing performance and enhancing  report build quality.  

Search Engine on Dancing (Web Application) (January 2024- May 2024)  

• Developed a responsive web application using React, ensuring an engaging user experience focused on  dancing.  

• Leveraged Python and NLTK libraries for information retrieval, enhancing natural language processing  and query expansion.  

• Implemented web crawling techniques to index 200,000 URLs, optimizing data retrieval for meaningful  search results.  

Optical Character Recognition for Devanagari script (February 2023- May 2023)  

• Developed a CNN model using TensorFlow and Keras, achieving 98% accuracy in detecting  handwritten characters.  

• Utilized Python and OpenCV for image preprocessing, refining 2,000 images of handwritten  Devanagari characters.  

• Employed Seaborn and scikit-learn for visualizing results, analyzing performance across 500  validation samples.  

CNN-Based Breast Cancer Detection Using Digital Mammogram (February 2022- June 2022)  

• Created a breast cancer tumor detection model using TensorFlow and Keras, achieving 86% accuracy  initially.  

• Enhanced the model over subsequent iterations, improving accuracy to 96% through rigorous training  methodologies.characters.  

• Developed an interactive Python GUI with Tkinter, allowing users to input data and visualize results  easily.

Performance Analysis of Breast Cancer Classification from Mammogram Images Using Vision Transformer, IEEE .Tech, 12/01/22, https://ieeexplore.ieee.org/document/10060315

Home Tutoring: Offered academic support in Mathematics and Science to middle and high school learners