🌟 5-Day Generative AI Intensive Course with Google 🌟

🚀 Overview

I am diving into the 5-Day Generative AI Intensive Course with Google, exploring the cutting-edge world of AI! This course covers everything from foundational LLMs to MLOps for generative AI. I'm completing each day's tasks, gathering insights, and sharing useful resources. Join me on this exciting learning journey!

---

🗓️ Daily Progress

Day 1: Foundational Large Language Models & Text Generation

📚 Assignments:

1. Optional Listen to the summary podcast for this unit.
2. Read the Foundational Large Language Models & Text Generation whitepaper.
3. Complete the Kaggle code lab on prompting fundamentals.

💡 What I Learned:

• Evolution of LLMs: From transformers to fine-tuning and inference acceleration.
• Fundamentals of prompt engineering for optimal interaction with LLMs.
• Hands-on experience with the Gemini API and prompt techniques.

---

Day 2: Embeddings and Vector Stores/Databases

📚 Assignments:

1. Optional Listen to the summary podcast for this unit.
2. Read the Embeddings and Vector Stores/Databases whitepaper.
3. Complete Kaggle code labs:
   • Build a RAG question-answering system.
   • Explore text similarity with embeddings.
   • Build a neural classification network with Keras.

💡 What I Learned:

• Concepts of embeddings and vector databases.
• Using embeddings for classifying and comparing textual data.
• How to bring live or specialist data into LLM applications.

---

Day 3: Generative AI Agents

📚 Assignments:

1. Optional Listen to the summary podcast for this unit.
2. Read the Generative AI Agents whitepaper.
3. Complete Kaggle code labs:
   • Talk to a database with function calling.
   • Build an agentic ordering system in LangGraph.

💡 What I Learned:

• Core components of AI agents and iterative development.
• Connecting LLMs to real-world systems.
• Building LangGraph agents for practical tasks.

---

Day 4: Domain-Specific LLMs

📚 Assignments:

1. Optional Listen to the summary podcast for this unit.
2. Read the Solving Domain-Specific Problems Using LLMs whitepaper.
3. Complete Kaggle code labs:
   • Use Google Search data in generation.
   • Fine-tune a Gemini model for a custom task.

💡 What I Learned:

• Building and applying specialized LLMs like SecLM and MedLM.
• Adding real-world data to models through grounding.
• Fine-tuning models for specific tasks using labeled data.

---

Day 5: MLOps for Generative AI

📚 Assignments:

1. Optional Listen to the summary podcast for this unit.
2. Read the MLOps for Generative AI whitepaper.
3. Explore the End-to-End Gen AI App Starter Pack.

💡 What I Learned:

• Adapting MLOps practices for generative AI.
• Leveraging Vertex AI tools for production-ready AI applications.

---

🎯 Final Task

Check out the bonus notebook for additional capabilities of the Gemini API, beyond what's covered in the course!

---

📂 Resources I’m Sharing

As I progress, I’m compiling all the whitepapers, code labs, and additional insights into a resource repository. Stay tuned for updates!

---

🙌 Let’s Connect

If you’re on a similar journey or want to collaborate, feel free to reach out or explore this repository. Together, let’s unlock the potential of Generative AI!