Curso Intensivo de IA Generativa de Google y Kaggle

Google y Kaggle lanzaron un curso intensivo de 5 días sobre IA generativa, el cual puedes tomar de forma completamente gratuita a través de plataformas abiertas como YouTube y Kaggle.

En el siguiente video describo de que trata el curso y más abajo puedes encontrar la lista de materiales de cada uno de los días:

@mkfnx Curso intensivo de #InteligenciaArtificial generativa por parte de #Kaggle y #Google #IA #IAGenerativa #GoogleGemini #MachineLearning ♬ sonido original - Miguel

Día 1

Introducción - “Foundational Large Language Models & Text Generation”:

• [Optional] Listen to the summary podcast episode (https://youtu.be/mQDlCZZsOyo) for this unit (created by NotebookLM, https://notebooklm.google.com/).
• Read the “Foundational Large Language Models & Text Generation” whitepaper (https://www.kaggle.com/whitepaper-foundational-llm-and-text-generation).

Unidad 1 - “Prompt Engineering”:

• [Optional] Listen to the summary podcast episode (https://youtu.be/F_hJ2Ey4BNc) for this unit (created by NotebookLM).
• Read the “Prompt Engineering” whitepaper (https://www.kaggle.com/whitepaper-prompt-engineering).
• Complete this code lab (https://www.kaggle.com/code/markishere/day-1-prompting) on Kaggle where you’ll learn prompting fundamentals. Make sure you phone verify (https://www.kaggle.com/settings) your account before starting, it’s necessary for the code labs.

💡What You’ll Learn Today you’ll explore the evolution of LLMs, from transformers to techniques like fine-tuning and inference acceleration. You’ll also get trained in the art of prompt engineering for optimal LLM interaction.

The code lab will walk you through getting started with the Gemini API and cover several prompt techniques and how different parameters impact the prompts.

Día 2

Unidad 2: “Embeddings and Vector Stores/Databases”:

• [Optional] Listen to the summary podcast episode (https://youtube.com/watch?v=1CC39K76Nqs) for this unit (created by NotebookLM, https://notebooklm.google.com/).
• Read the “Embeddings and Vector Stores/Databases” whitepaper (https://kaggle.com/whitepaper-embeddings-and-vector-stores).
• Complete these code labs on Kaggle:
  • Build a RAG question-answering system over custom documents - https://www.kaggle.com/code/markishere/day-2-document-q-a-with-rag
  • Explore text similarity with embeddings - https://www.kaggle.com/code/markishere/day-2-embeddings-and-similarity-scores
  • Build a neural classification network with Keras using embeddings - https://www.kaggle.com/code/markishere/day-2-classifying-embeddings-with-keras

Día 3

Unidad 3: “Generative AI Agents”:

• [Optional] Listen to the summary podcast episode (https://youtu.be/H4gZd4BCrDQ) for this unit (created by NotebookLM).
• Read the “Generative AI Agents” whitepaper (https://www.kaggle.com/whitepaper-agents).
• Complete these code labs on Kaggle: Talk to a database with function calling - https://www.kaggle.com/code/markishere/day-3-function-calling-with-the-gemini-api Build an agentic ordering system in LangGraph - https://www.kaggle.com/code/markishere/day-3-building-an-agent-with-langgraph/

Día 4

Unidad 4: “Domain-Specific LLMs”:

• [Optional] Listen to the summary podcast episode (https://youtu.be/b1a4ZOQ8XdI) for this unit (created by NotebookLM).
• Read the “Solving Domain-Specific Problems Using LLMs” whitepaper - https://www.kaggle.com/whitepaper-solving-domains-specific-problems-using-llms
• Complete these code labs on Kaggle:
  • [Optional] Use Google Search data in generation. (Note: Grounding with Google Search has been released as a limited launch and is not available in all locations. The EEA, UK, and CH regions will be supported at a later date) - https://www.kaggle.com/code/markishere/day-4-google-search-grounding
  • Tune a Gemini model for a custom task - https://www.kaggle.com/code/markishere/day-4-fine-tuning-a-custom-model

Día 5

Unidad 5: “MLOps for Generative AI”:

• [Optional] Listen to the summary podcast episode (https://youtu.be/k9S6IhiUUj4) for this unit (created by NotebookLM, https://notebooklm.google/).
• Read the “MLOps for Generative AI” whitepaper - https://www.kaggle.com/whitepaper-operationalizing-generative-ai-on-vertex-ai-using-mlops
• No code lab for today! We will do a code walkthrough and live demo of goo.gle/e2e-gen-ai-app-starter-pack (https://goo.gle/e2e-gen-ai-app-starter-pack), a resource created for making MLOps for Gen AI easier and accelerating the path to production.

Bonus

Codelab “Extra API features to try”

https://www.kaggle.com/code/markishere/bonus-day-extra-api-features-to-try/

Grabación Lives Preguntas y Respuestas

https://www.youtube.com/playlist?list=PLqFaTIg4myu-b1PlxitQdY0UYIbys-2es