Material

📚

Session 1: Introduction to LLM on Production ✨

• Overview of LLM (Large Language Model) Applications in Production

• Importance of LLM in Various Industries

• Challenges and Considerations in Deploying LLM Models

Session 2: LLM Managed Service (Claude-3, OpenAI) ✨

• Introduction to Claude-3 and Other Managed LLM Services

• Comparison of Features and Pricing

• Use Cases and Benefits of Using Managed LLM Services

Session 3: Building RAG Using Vector Database ✨

• Understanding RAG (Retrieval-Augmented Generation) Models

• Introduction to Vector Databases PgVector

• Practical Implementation of RAG Using a Vector Database

Session 4: Handling with Open Source LLM (Loading, Fine-Tuning, etc.) ✨

• Overview of Popular Open-Source LLM Frameworks (e.g., Hugging Face Transformers, LLAMA)

• Loading Pre-Trained Models and Datasets

• Techniques for Fine-Tuning LLMs for Specific Tasks

Session 5: Optimizing Your Open Source LLM Deployment ✨

• Best Practices for Deploying LLMs in Production Environments

• Performance Optimization Tips (e.g., Batch Size, Hardware Acceleration, Quantization)

Session 6: Online vs Offline Metrics for LLM Applications ✨

• Understanding Metrics for Evaluating LLM Performance in Online vs. Offline Scenarios

• Key Metrics: Accuracy, Latency, Throughput, Cost-Efficiency

Session 7: Building RAG Using Open Source LLM ✨

• Step-by-Step Guide to Building Retrieval-Augmented Generation Models Using Open-Source LLM Frameworks

• Integrating with External Data Sources for Retrieval

• Advanced Techniques and Extensions for RAG Models