A comprehensive portfolio showcasing my work in Large Language Models (LLM) and Natural Language Processing. This repository contains examples ranging from fundamental ML concepts to advanced NLP applications.

This portfolio demonstrates practical applications of modern NLP techniques and LLM technologies through hands-on projects. Each component is designed to showcase different aspects of AI development, from basic concepts to production-ready systems.

• Transformer Architecture: Implementation and explanation of core transformer components
• Tokenization: Text preprocessing and tokenization techniques
• Sentiment Analysis: Basic emotion detection using pre-trained models
• LLM Terminology: Comprehensive guide to essential concepts

• Framework Integration: Advanced problem-solving using LangChain
• OpenAI API: Seamless integration with GPT models
• Reasoning Chains: Implementation of complex multi-step reasoning

• Language Support: Story generation in Turkish, English, German, French, and Spanish
• Quality Assessment: Automated story quality evaluation
• Sentiment Analysis: Emotion detection in generated content
• Interactive Interface: User-friendly story creation system

• BERT Integration: Custom BERT-based chatbot for Turkish customer service
• Web Interface: Professional Gradio-powered web application
• Real-time Processing: Instant response generation

• FAISS Vector Store: High-performance document indexing
• PDF Processing: Automatic document ingestion and processing
• Natural Language Queries: Conversational question-answering interface
• Real-time Retrieval: Instant information extraction from documents

• FastAPI Backend: High-performance API server
• Streamlit Frontend: Interactive web interface
• Code Analysis: Automated code review and suggestions
• Security Scanning: Built-in security vulnerability detection
• Docker Ready: One-command deployment

transformers, torch, langchain, openai, gradio, streamlit, fastapi

• Models: GPT-4, BERT, RoBERTa
• Vector Store: FAISS
• Embeddings: OpenAI Embeddings
• Languages: Python, JavaScript

pip install -r requirements.txt

Create a .env file in the root directory:

OPENAI_API_KEY=your_api_key_here

Run all applications with a single command:

docker-compose up -d

Navigate to specific modules and explore:

cd 01-LLM  # Start with fundamentals
python examples/tokenization_demo.py

cd ../04-LLM  # Try vector search
python semantic_search_example.py

cd ../07-LLM  # Deploy applications
docker-compose up -d

• 🌍 Multilingual Support: Generate and analyze content in 5+ languages
• 🤖 Intelligent Chatbot: Turkish-optimized customer service bot
• 📖 Document Q&A: Query PDF documents using natural language
• 🔒 Code Security: Automated security analysis for code projects
• 🚀 Easy Deployment: Docker-containerized for seamless deployment

├── 01_llm_fundamentals/     # Basic LLM concepts and implementations
├── 02_langchain_projects/   # LangChain framework applications
├── 03_story_generation/     # Multilingual story creation system
├── 04_gradio_deployment/    # Web-based chatbot deployment
├── 05_rag_system/          # Retrieval Augmented Generation
├── 06_ai_code_assistant/   # Dockerized code analysis tool
├── docker-compose.yml      # Multi-container deployment
├── requirements.txt        # Python dependencies
└── README.md              # This file

This portfolio is perfect for those looking to gain hands-on experience in:

• Transformer Architecture: Deep understanding of attention mechanisms
• LangChain Framework: Building complex AI workflows
• RAG Systems: Implementing retrieval-augmented generation
• Multilingual NLP: Working with multiple languages
• Production Deployment: Docker containerization and web deployment
• API Integration: OpenAI and other ML service integrations