Build Serverless AI on Google Cloud Platform – Fast & Practical

In-depth project series for developers who want portfolio-ready AI demos using GCP.

Are you a developer looking to solve real-world problems using AI/ML with Google Cloud Platform?

We offer a hands‑on & interactive experience: Build, deploy, and demo live AI agentic systems on GCP.

    Upcoming course:
  • - Serverless Agents with Next.js and GCP

We focus on solving practical problems with deployable outcomes using real-world use cases.

You will walk away with portfolio and employer-ready confidence in the fast changing world of AI.

Register your interest today and you will be the first to get early access.

Serverless Agents with Next.js and Google Cloud Platform

Deep-dive project guiding you step-by-step to build a comprehensive, real-world AI customer support system on Google Cloud Platform.

Course Syllabus

Foundations

  • Course Intro: What We’ll Build
  • Stack Overview: Next.js, @ai-sdk/google, Gemini, Cloud Run
  • GCP Setup: Project, Services (Cloud Run, Artifact Registry, Secrets and more)
  • GCP Setup 2: Budgets and Alarms
  • Scaffolding: Next.js App & Project Structure
  • AI SDK Integration: Install AI SDK & Connect to Gemini

Basic Agent

  • First Chat: Chat API Route with streamText
  • Simple Chat UI: Chat UI (streaming) in Next.js
  • Containerize: Next.js: Dockerfile + local test
  • Deploy: Deploy basic Agent to Cloud Run

Agent With Memory and Tools

  • Memory: Conversation Memory (client state)
  • Persistence: Persist Sessions to Firestore
  • Tools: Function Calling Basics
  • Tool 1: Order Lookup (mock)
  • Tool 2: RAG for frequently accessed questions
  • Tools UI: Render Tool Responses in the Chat UI
  • UI Polish: Typing indicator, message shapes, loading states
  • Evals: Evals with Evalite (AI SDK)

Secure & production-ready Agent

  • Auth: Setup Authentication
  • User Context: Context in Prompts + RBAC Responses
  • Rate Limits: Rate Limiting/Quotas
  • Error Handling: Cover the unhappy paths
  • Guardrails: Input/Output filtering + safety system prompts
  • Observability: Cloud Logging + Cloud Trace on Cloud Run
  • Observability 2: Agent Observability with Langfuse
  • Evals: Create a test set + red-team prompts; pass/fail gates

Enterprise-ready & Multimodal Agent

  • Analytics: Export logs to BigQuery
  • Multimodal: Support Images in chats
  • Cost & Performance: Concurrency, min instances, autoscaling tuning
  • Conclusion: Final Walkthrough & What to Ship Next

Here’s why this program is unlike traditional courses:

  • Hands-on Projects: You'll work through structured projects to build real AI applications using modern techniques and best practices.
  • End-to-End Mastery: Cover every step from data handling to production deployment and monitoring
  • Industry-standard best practices: Build scalable, efficient, and maintainable AI systems.
  • Expert Mentoring: Continuous support from mentors and an active peer community.

This series will transform your AI capabilities by immersing you in realistic workflows, empowering you to confidently build solutions that industry demands.

Who Is This Program For?

This is a hands-on program for people willing to put in the work to build skills with real-world impact.

This program is ideal for software engineers, researchers, data scientists, data analysts, technical managers, and anyone ready to apply AI to real-world challenges.

Prerequisites for Success:

  • Experience writing code (JavaScript preferred, but any language is fine).
  • Some familiarity with GCP services and building simple APIs. Docker experience is beneficial but not required.
  • Comfortable asking questions, sharing progress, and supporting peers.
  • Prepared to dedicate time and effort to building impactful AI skills.