Welcome to my GitHub! I'm Davut Bayık, a passionate data scientist, machine learning enthusiast, and aspiring AI expert. Here's where I showcase my projects, research, and contributions. Feel free to explore and collaborate!
I am a Data Scientist, and I specialize in:
- Multi Agent AI frameworks and RAG orchestration
- Aritifical Intelligence API's and prompt engineering
- Machine learning and deep learning model building/deployment
- Data preprocessing and cleaning
- Data visualization
- Predictive modeling
- Natural Language Processing (NLP)
- Computer Vision
- Desktop and Web Applications using Python frameworks
- Full stack web development
I’m currently working on data science projects related to building AI agents, ML algorithms, data analysis, interactive dashboards and PyQt desktop apps.
- Programming Languages: Python, SQL, Javascript, Php, HTML/CSS
- Frameworks & Libraries: CrewAI, Langchain, Langgraph, PyTorch, Tensorflow, Scikit-learn, Pandas, Numpy, Streamlit, Flask, FastAPI, FPDF, Xlswriter, PyQt6, Tkinter, PyGame, Tailwindcss, React
- Tools & Platforms: GitHub, Jupyter, VS Code, Docker, AWS, Google Cloud
- Databases: PostgreSQL, MySQL, MsSQL, MongoDB
Here are some of my favorite projects:
- Description: AutoAdvisor is an AI-powered business strategy assistant that validates user-submitted business ideas through a web application, generates strategic reports using multi-agent reasoning, and delivers actionable insights including a SWOT analysis.
- Technologies: CrewAI, Langchain, OpenAI API, SerpAPI, Streamlit
- Goal: Help users transform raw business ideas into validated, actionable strategies by leveraging AI agents for analysis, market research, and strategic planning.
- Description: RAG Chatbot is an AI-powered assistant that enables users to upload documents and ask natural language questions about their contents. Leveraging Retrieval-Augmented Generation (RAG), the chatbot extracts relevant context from the documents and provides intelligent, context-aware answers through a user-friendly custom Whatsapp inspired chat interface.
- Technologies: LangChain, OpenAI API, Google Gemini API, FAISS, Streamlit, PyPDF
- Goal: Empower users to interact with unstructured documents using conversational AI by combining retrieval and generation techniques, making document exploration faster, smarter, and more intuitive.
- Description: A Python-based automation tool that uses Selenium to send personalized WhatsApp messages through WhatsApp Web by simulating user interactions in a browser.
- Technologies: Python, Selenium WebDriver, Chromedriver.
- Goal: To streamline and automate the process of sending WhatsApp messages to individual or multiple contacts without manual intervention.
- Description: A sentiment analysis model that classifies metacritic reviews text data into positive, neutral, or negative sentiments using a fine-tuned BERT model.
- Technologies: Python, BERT, Hugging Face, NLTK
- Goal: To analyze user and critic reviews from metacritic games for insights into games sentiment.
- Description: A custom OOP class for scraping metacritic games, movies and tv shows metadata and reviews from metacritic's backend API using requests.
- Technologies: Python, Requests, Pandas
- Goal: To efficiently extract structured, high-quality metadata and reviews for games, movies, and TV shows from Metacritic’s backend API—enabling deep analysis, sentiment mining, and recommendation system research without the overhead of frontend scraping.
- Description: A Streamlit dashboard and FastAPI endpoint that predicts whether a user will purchase a product based on their gender, age, and estimated salary.
- Technologies: Python, Streamlit, FastAPI, Scikit-learn
- Goal: To help businesses optimize their digital ad targeting using machine learning models.
- Description: A machine learning model that predicts food delivery times based on time of the delivery, distance, courier's rating and age, current weather condition, traffic and various factors.
- Technologies: Python, Scikit-learn, Pandas, Seaborn
- Goal: To help food delivery companies to optimize their resource allocation and delivery strategy.
Feel free to check out my repositories and let me know if you want to collaborate or have any questions. Happy coding! 👨💻
If you're interested in working together or discussing projects, don't hesitate to reach out. I'm always open to learning and collaborating on exciting projects!