Sai Kumar Adapa

Introduction:

"TeckyBot" is a web research tool designed to empower users to ask questions about websites and news articles by providing the URLs of the pages.

Features:

1. Dynamic Input Handling: Allows users to input multiple URLs dynamically through the Streamlit interface.
2. Question Answering: Employs a retrieval-based approach to answer user queries based on the information extracted from the provided URLs.
3. User Interface (Streamlit): Implements a user-friendly web interface using Streamlit for easy interaction and seamless user experience.
4. Flexible Prompting System: Utilizes customizable prompting templates for question answering, allowing flexibility in the types of queries users can make.

Techstack :

• Language: Python
• Web Scraping: requests, BeautifulSoup
• Libraries & Models: langchain, HuggingFaceEmbeddings, GooglePalm
• Vector Database: FAISS
• UI: streamlit

Introduction:

Introducing my versatile project - a smart query resolver for CSV data with built-in chart plotting capabilities. Simplify data exploration and visualization effortlessly!.

Features:

1. Dynamic Input Handling: Allows users to input CSV files dynamically through the Streamlit interface.
2. Question Answering: Employs a retrieval-based approach to answer user queries based on the information extracted from the provided URLs.
3. User Interface (Streamlit): Implements a user-friendly web interface using Streamlit for easy interaction and seamless user experience.
4. Flexible Prompting System: Utilizes customizable prompting templates for question answering, allowing flexibility in the types of queries users can make.

Technologies Used:

• Language: Python
• Libraries & Models: langchain, pandasai
• Vector Database: FAISS
• UI: streamlit

Introduction:

Empower your PDF interactions! My application leverages AI to swiftly answer your document queries, enhancing search and comprehension effortlessly.

Features:

1. Dynamic Input Handling: Allows users to input PDFs dynamically through the Streamlit interface.
2. Question-Answering: Users can ask questions related to the uploaded PDF.
3. User Interface (Streamlit): Implements a user-friendly web interface using Streamlit for easy interaction and seamless user experience.
4. Flexible Prompting System: Utilizes customizable prompting templates for question answering, allowing flexibility in the types of queries users can make.

Technologies Used:

• PyPDF2 : Used for reading PDF files and extracting text content
• LangChain : A custom library for handling natural language processing tasks, including vectorization, embeddings, and language models
• FAISS : A library for efficient similarity search and clustering of dense vectors
• pandas: Used for organizing and structuring extracted data in a tabular format
• Streamlit: Used for creating interactive and user-friendly applications

Tech Stack: OpenAI APIs, Streamlit, Python, Pandas

Introduction:

Unlock financial insights from news articles! This project focuses on extracting key financial data such as Company Name, Stock Symbol, Revenue, Net Income, and EPS from a diverse range of news articles. Experience the power of this automated financial data extraction tool to enhance the work of financial analysts and streamline the information gathering process.

Features:

1. Automatic Data Extraction: Seamlessly extract financial data from news articles without manual intervention.
2. Key Metrics: Retrieve essential financial metrics, including company name, stock symbol, revenue, net income, and earnings per share (EPS).
3. NLP-Powered Analysis: Leverage Natural Language Processing to comprehend and analyze complex textual information.
4. Data Structure: Organize extracted data in a structured format for easy analysis and interpretation.
5. Adaptability: The system is adaptable to a variety of news articles and sources, ensuring reliable data extraction across different contexts.
6. Time Efficiency: Streamline the process of financial analysis by automating the extraction of crucial data points from textual content.
7. Insightful Reporting: Generate insightful reports based on the extracted financial data, aiding in decision-making processes.
8. Integration Possibilities: Explore opportunities to integrate the system with financial analysis tools or databases for enhanced functionality.

Technologies Used:

• OpenAI API: Utilized for natural language understanding and content extraction.
• Python: The core programming language behind the project's logic and data processing.
• Pandas: Used for organizing and structuring extracted data in a tabular format.

Tech Stack: Pydantic Model, OpenAI APIs, Streamlit, Python, BeautifulSoup

Introduction:

project is a data extraction and analysis tool that leverages OpenAI's GPT-3.5-turbo language model and Pydantic for efficient and accurate processing of film-related information. The primary goal of this project is to extract relevant data from a Wikipedia page containing details about the film data. The extracted data is then filtered based on user-defined constraints, providing a clean and structured JSON output for API uses.

Features:

• Data Extraction: Extract data from web pages using OpenAI models.
• Filtering: Implement filtering based on user-specified criteria.
• Real-time Interaction: Provide a user-friendly interface for real-time interaction.
• Data Validation: Utilizing the Pydantic model to enforce strict data types for accessing the data.

Technologies Used:

• OpenAI API: Leveraged for advanced language processing and data extraction.
• Pydantic: Implemented for data validation and modeling.
• Streamlit: Used for creating interactive and user-friendly applications.
• Python: The core programming language behind the project's logic.
• BeautifulSoup: Employed for web scraping and data extraction.

Tech Stack: OpenAI APIs, Python, Pandas, MySQL, Streamlit

Introduction:

The College Question & Answering System with General Knowledge is a sophisticated tool designed to provide insightful answers to queries related to college-related information. Leveraging the power of OpenAI's GPT-3.5-turbo model, this system can process a wide range of questions related to student GPA(marks), fees, and Aggregations (Avg, Max, Min, Total).

Features:

1. Dynamic Query Processing: The system excels in dynamically processing user queries related to college data, delivering comprehensive and context-aware responses.
2. Database Integration: Seamlessly integrated with MySQL, ensuring efficient storage and retrieval of student records for quick and reliable access to relevant data.
3. GPA and Due Fees Calculation: Empowered to calculate GPAs for specific students or aggregate GPAs, such as average, minimum, or maximum, for a given semester. Additionally, it provides detailed information about individual students' fees or the total fees for an entire semester, including paid, pending, or total fees.
4. General Knowledge Questions: The system is adept at solving general knowledge questions, broadening its utility beyond college-specific inquiries.
5. User-Friendly Interface: Designed with a user-friendly interface powered by Streamlit, allowing users to input questions effortlessly and receive clear and concise answers.
6. Extensible Functions: The system's flexibility allows for the addition of new functions, addressing diverse aspects of college-related queries and enhancing its adaptability.

Technologies Used:

• OpenAI API and OpenAI Functions: Empowering the system with advanced language processing capabilities for dynamic query interpretation.
• Python: The core programming language responsible for system logic, integration, and interaction with OpenAI's language model.
• MySQL: Serving as the database management system for efficient storage and retrieval of student-related information.
• Streamlit: Providing an interactive web interface, allowing users to input questions and receive responses with ease.

Uses:

This tool proves invaluable for students, faculty, and administrators seeking quick and accurate information about academic performance, fees, and general knowledge related to college operations. Its dynamic processing and extensible functions make it adaptable to evolving requirements in the educational domain.

Introduction:

The project is an intelligent question and answering system designed for online course platforms. It provides accurate responses to user queries, enhancing the learning and customer satisfaction experience.

Features:

1. Reduce Efforts: Reduces the need for manual intervention in responding to frequently asked questions, saving time for platform maintainers.
2. Time Savings: Quick resolution of queries saves users time, enabling them to focus more on their learning objectives.
3. Increased Satisfaction: Users benefit from efficient support, leading to higher satisfaction levels with the online course platform.
4. User Interface (Streamlit): Implements a user-friendly web interface using Streamlit for easy interaction and seamless user experience.

Techstack :

• Language: Python
• data Manipulation: MySQL
• Libraries & Models: langchain, HuggingFaceEmbeddings, GooglePalm
• Vector Database: FAISS
• UI: streamlit

Tech Stack: Computer Vision, MediaPipe, Python, OpenCV

Introduction:

Jump into the boxing ring without any controllers! This two-player boxing game brings the thrill of boxing matches, all controlled by human gestures. Using advanced computer vision techniques, players can throw punches, dodge, and master the art of boxing, all in real-time.

Features:

1. Real-time Gesture Detection: Play the game with real human body gestures, no controllers required.
2. Two-Player Combat: Challenge a friend and have boxing matches right in front of your computer.
3. Visual Feedback: Know every move with instant feedback on the video feed, highlighting player moves and scores.
4. Customizable Settings: Tailor the game to your preference. Adjust settings, switch modes, and curate your own boxing experience.
5. Scalability: The game isn't just limited to its current form. It has the potential to recognize more gestures, introduce more players, or even integrate new gaming scenarios.
6. Minimal Setup: No need for fancy gear. With a basic setup and your computer, you're good to go, making it accessible for a wide audience.
7. Robustness: Be it a bright sunny room or a dimly lit space, variations in lighting don't deter the game. It's also attuned to understand and adapt to different player builds and postures.
8. Safety First: We care about you. By relying on non-contact gestures, the game ensures a fun experience without any physical risks.
9. Platform Flexibility: Not just confined to a PC, the underlying technology is flexible and can be ported to different platforms or integrated with other gaming systems for a broader experience.

Technologies Used:

• OpenCV: For capturing and processing video frames, making the gameplay smooth and responsive.
• MediaPipe: Utilized for its pre-trained models, it aids in accurate gesture and pose estimation.
• Python: The core programming language behind the game's logic and mechanics.

Github README File: Click Here

Tech Stack: Computer Vision, Python, OpenCV, cvzone

Introduction:

Ever found yourself meticulously counting coins? With the Money Counter using Computer Vision, that's a thing of the past. The project simplifies counting by differentiating coin denominations and calculating the total value seamlessly.

Features:

1. Real-time Coin Detection: Swiftly detects and differentiates coins.
2. Multi-denomination Support: Compatible with multiple denominations for a comprehensive count.
3. Total Value Calculation: Instantly get the overall monetary value of the detected coins.
4. Visual Feedback: View detected coins and their values in real-time.
5. Stability in Counting: Ensures precision in counting even with a plethora of coins.

Technologies Used:

• OpenCV: Utilized for capturing and processing images, vital for the coin detection mechanism.
• cvzone: Streamlines contour handling and image stacking, enhancing the efficiency of the detection process.
• Python: The backbone programming language driving the project's logic and functionalities.

Live Link : Click Here

Tech Stack: HTML, CSS, JavaScript, Bootstrap

Introduction:

A comprehensive portfolio detailing Sai Kumar Adapa's academic accomplishments, skills, interests, certifications, and professional experiences. The website offers a clean, intuitive interface, enabling easy navigation for users and a smooth experience. The site is developed using a combination of HTML, CSS, JavaScript, and other frameworks to ensure it's responsive and visually appealing.

Features:

1. Responsive Design: Optimized for various device compatibilities.
2. Dynamic Visual Feedback: Includes animations and transitions for user engagement.
3. Interactive Sections: Detailed sections about Sai Kumar's education, experience, and interests.
4. Integrated Social Links: Direct links to Sai Kumar's professional profiles and contact methods.
5. Demo Videos: Embedded demo videos for each project to provide a comprehensive understanding of the functionalities and features.

Technologies Used:

• HTML: For structuring the content on the website.
• CSS: To style and design the website's appearance.
• JavaScript: Adds interactivity and dynamic behavior to the site.
• Bootstrap: A responsive design framework used for multi-device compatibility.