My Learnings from Day 1a & 1b of the Google AI Agents Intensive

Pavansai Ponguluri — Mon, 10 Nov 2025 18:04:11 +0000

Introduction

Hi everyone!
I’m currently taking part in the Google × Kaggle AI Agents Intensive, and I’ve completed Day 1a (Agent Basics) and Day 1b (Agent Architectures).
This post is a short reflection on what I learned, what I built, and how it went!

Day 1a – Agent Basics

Learned what an AI agent is and how it can automate reasoning and decision-making.

Tried my first simple agent using prompts and instructions.

Learned how to safely manage API keys and secrets in Kaggle.

Saw how agents can follow instructions to perform tasks step-by-step.

Day 1b – Agent Architectures

Explored how multiple agents can work together in a multi-agent system.

Learned about roles such as Coordinator, Planner, and Executor.

Built sequential and parallel workflows to coordinate agents.

Encountered a “Resource Exhausted” error — and learned that it happens when the API quota is used up.

Understood how rate limits and quotas affect API-based systems.

Key Takeaways

AI agents can make complex tasks easier by breaking them into smaller steps.

Good coordination is essential when multiple agents share information.

Resource and rate-limit management is an important part of building reliable systems.

The Kaggle notebooks made it easy to learn, run, and visualize results interactively.

DEV Community: Pavansai Ponguluri

My Learnings from Day 1a & 1b of the Google AI Agents Intensive