abhaypawar / Oracle_GenAI

Generative AI by Oracle Cloud Infrastructure

Oracle_GenAI

Generative AI by Oracle Cloud Infrastructure

Fundamentals of Large Language Models (LLMs)

LLM Basics

• Introduction to LLMs: Overview, significance, and use cases.
• Evolution of LLMs: Historical perspective, key milestones, and breakthroughs.
• Key Concepts: Tokens, embeddings, attention mechanisms.

LLM Architectures

• Transformer Architecture: Understanding the backbone of modern LLMs.
• Variations and Enhancements: BERT, GPT, T5, and other notable architectures.
• Comparative Analysis: Strengths, weaknesses, and ideal applications.

Prompt Engineering

• Prompt Design: Best practices, examples, and common pitfalls.
• Prompt Optimization: Techniques for refining and improving prompts.
• Real-world Applications: Case studies and examples of effective prompt engineering.

Fine-Tuning Techniques

• Transfer Learning: Adapting pretrained models to new tasks.
• Few-Shot and Zero-Shot Learning: Leveraging minimal data for fine-tuning.
• Advanced Fine-Tuning Methods: Techniques like T-Few and others.

Fundamentals of Code Models

• Introduction to Code Models: Understanding models designed for code generation and analysis.
• Applications: Code completion, bug detection, code translation.
• Popular Code Models: Codex, GPT-3 for code, and others.

Multi-modal LLMs and Language Agents

• Multi-modal LLMs: Integrating text, image, and other modalities.
• Language Agents: Creating agents that interact with users through natural language.
• Applications: Real-world use cases of multi-modal LLMs and language agents.

---

OCI Generative AI Deep-Dive

Pretrained Foundational Models

• Generation: Models focused on text generation.
• Summarization: Models designed for text summarization.
• Embedding: Understanding embedding models and their applications.

Flexible Fine-Tuning

• T-Few Technique: Detailed exploration of the T-Few fine-tuning approach.
• Custom Fine-Tuning: Techniques for tailoring models to specific tasks.

Model Inference

• Inference Techniques: Methods for efficient and effective model inference.
• Optimizing Inference: Strategies for reducing latency and improving performance.

Dedicated AI Clusters

• Setting Up AI Clusters: Best practices for deploying AI clusters.
• Scalability and Performance: Ensuring optimal performance and scalability.

Generative AI Security Architecture

• Security Considerations: Key security aspects in generative AI.
• Best Practices: Implementing robust security measures in AI deployments.

---

Build a Conversational Chatbot with OCI Generative AI

Understand RAG (Retrieval-Augmented Generation)

• Concept of RAG: Integrating retrieval with generation for improved responses.
• Applications: Use cases and examples of RAG in chatbots.

Vector Databases

• Introduction to Vector Databases: Understanding the role of vector databases in AI.
• Popular Vector Databases: Overview of key vector database solutions.

Semantic Search

• Concept of Semantic Search: Going beyond keyword-based search.
• Implementing Semantic Search: Best practices and techniques.

Build Chatbot Using LangChain Framework

• LangChain Framework Overview: Introduction to LangChain.
• Prompts: Crafting and optimizing prompts for chatbots.
• Models: Selecting and fine-tuning models within LangChain.
• Memory: Implementing memory for stateful interactions.
• Chains: Creating and managing conversation chains.

Trace and Evaluate Chatbot

• Debugging and Tracing: Techniques for troubleshooting chatbot issues.
• Evaluation Metrics: Key metrics for assessing chatbot performance.

Deploy on OCI

• Deployment Strategies: Best practices for deploying chatbots on OCI.
• Scaling and Maintenance: Ensuring your chatbot remains scalable and maintainable.