Mollie Pettit (she/her)

AI Dev Rel @ Google; doesn't eat animals🐣; RA poly queer🌈; dancer💃
Find me on BlueSky:

[Mollie Pettit (she/her) @MollzMP]

Dear friends,

Find me on BlueSky: https://t.co/WjxIK4hpnB

Thanks! 🤗

[Google Cloud Tech @GoogleCloudTech]

A single AI agent can't be an expert in everything. Our new labs show how to orchestrate multi-agent systems where specialized agents collaborate to solve complex problems.

Learn to use the MCP as a router for tasks and enable direct A2A communication ↓ https://t.co/IFqMc01Vkj

[Google Cloud Tech @GoogleCloudTech]

So you've built an AI Agent. Now, learn how to deploy it with our new set of hands-on labs.

You'll get hands-on experience deploying your agent with:
⚙️ Agent Engine
🚀 Cloud Run
☸️ GKE

Learn more and grab the links to all three labs, here ↓ https://t.co/x65AAnxIC6

[Google Cloud Tech @GoogleCloudTech]

Build a production-ready AI security foundation with three news labs.

Get hands-on experience with:
🛡️ Securing your training data
🔒 Hardening the AI infrastructure
🕵️ Controlling inputs and outputs to prevent abuse

Check them out ↓ https://t.co/BmorMPTBIG

[Google Cloud Tech @GoogleCloudTech]

How do you evaluate if your gen AI application is actually good?

Our new hands-on labs are designed to help you master AI evaluation—from single prompts and RAG systems to complex AI agents. Check out this blog for more info and links to all four labs ↓ https://t.co/9JXkdFXClT

[Richard Seroter @rseroter]

This was so easy, it's almost embarrassing.

With one prompt, can I get @antigravity to design and build a data schema for a multi-tenant SaaS billing system?

Yes. And it uses its browser, agents, and terminal to do it.

Try it yourself. It's awesome.
https://t.co/UUp0lO3Gbf

[Google Open Source @GoogleOSS]

Open standards for agentic AI, like MCP and A2A, are great to have and even more so if you know how to use them. Check out this blog post with some labs to get you started.

[Mollie Pettit (she/her) @MollzMP]

For those who want to dive straight into the code, here are the direct links to the labs. 🤓

- Getting Started with MCP, ADK and A2A 👉https://t.co/L0MHBcz5Nv
- MCP Toolbox for BigQuery Databases 👉https://t.co/uL2ijQTTJ6
- Build a Travel Agent using MCP Toolbox 👉https://t.co/RzHZJpxIE9

#HandsOnLab #AIAgents

[Mollie Pettit (she/her) @MollzMP]

A single AI agent, no matter how powerful, can't be an expert in everything.

To help you build more sophisticated multi-agent systems, I'm sharing a blog post that introduces hands-on labs from my colleagues @JackWoth98 & @iRomin. 🧵👇
https://t.co/CdNW78UP9o

@GoogleCloudTech #ProductionReadyAI

[Mollie Pettit (she/her) @MollzMP]

In these labs, you'll get hands-on experience facilitating communication between #AIAgents by:

🤝 Implementing the Agent-to-Agent (#A2A) Protocol
📡 Using the Model Context Protocol (#MCP) as a router for tasks

[Shir Meir Lador @shirmeir86]

As a data scientist focused on SFT, I’ve always wondered where RL truly fits in and where do TPUs really shine.
Joining Google finally gave me the answers!
Level up your AI strategy with our latest #AgentFactory recap:

https://t.co/X0jSUX5jsR

[Mollie Pettit (she/her) @MollzMP]

For those who want to dive straight into the code, here are the direct links to the labs. 🤓

- Evaluate Single LLM Outputs: https://t.co/hEzUHq5jKp
- Evaluate RAG: https://t.co/EGPLstaobg
- Evaluating Agents: https://t.co/PWUBvRU8ZU
- Evaluate BigQuery Agents: https://t.co/CNfnywa9UO

#HandsOnLab

[Mollie Pettit (she/her) @MollzMP]

How do you know your #GenAI app is actually good? A "vibe check" isn't enough for production.

To help, my colleagues @SmithaKolan, @anniewangtech, & Rachael Deacon-Smith created 4 new hands-on labs to help you master #AIEvaluation.

This blog post introduces them all. ���
https://t.co/PQ4dzgIBdB

@GoogleCloudTech

[Mollie Pettit (she/her) @MollzMP]

In these labs, you'll get hands-on with:

📏 Evaluating Single Prompts with Vertex AI
🔎 Evaluating RAG Systems with Vertex AI
📈 Evaluating Agents with ADK
🤖 Evaluating BigQuery Agents with ADK and GenAI Eval Service

#HandsOnLab #ProductionReadyAI

[Richard Seroter @rseroter]

You have a genuinely dumb idea for an app. Will your favorite AI tool build it anyway?

I wondered if @antigravity would. It did, but also helped me go back, refine my thinking, and land on something useful.

https://t.co/Txntq5N8qu

[Agentic Amit @agenticamit]

Happy New Y{AI}r! 🥂

I believe this is the year Multi-Agent Systems become more popular

I recently tried to build a Deep Research tool and realized that a single agent is… slow.

So I built a squad. Here’s how I cut latency by 60% with #GoogleADK & #CloudRun. 🧵👇

[Mollie Pettit (she/her) @MollzMP]

Looking for a New Year's Resolution? 📝 You couuuuld make 2026 the year you learn production-ready #AI.

To help you with this, I'm resharing the Production-Ready AI learning path, a self-paced curriculum to take your #GenAI skills to the next level.

Start your resolution here 🤗🤓🧑‍💻👇
https://t.co/PgsKs0hELC

@GoogleCloudTech #LearnToCode

[Mollie Pettit (she/her) @MollzMP]

Get started building #AIAgents today! Labs by @iamthuya and myself. :D

#ProductionReadyAI #HandsOnLab #ADK

[Google Cloud Tech @GoogleCloudTech]

Ready to build AI that can reason, plan, and act? Easily build your own AI agents, even without deep expertise, thanks to Agent Development Kit (ADK)!

This blog introduces three hands-on labs that cover the core patterns of agent development → https://t.co/2sifPDkHa7

[Mollie Pettit (she/her) @MollzMP]

@GoogleCloudTech P.S. If anyone has suggestions for new topics or improvements, please let us know in this survey 🤗

https://t.co/08mHk7sCoH

[Mollie Pettit (she/her) @MollzMP]

P.S. If you have suggestions for new topics or improvements, let us know in this survey: https://t.co/08mHk7sCoH

[Mollie Pettit (she/her) @MollzMP]

A massive amount of work went into this from my colleagues and me to create a practical learning path for production AI.

From building your first #AIAgent to securing it and deploying it to the cloud. Hope you find it useful!

#ProductionReadyAI #AIApps

[Google Cloud Tech @GoogleCloudTech]

Introducing the Production-Ready AI with Google Cloud Learning Path, a free series designed to take your AI projects from prototype to production.

The path's curriculum combines Gemini models with #VertexAI, GKE, and more. Check out the modules, here ↓ https://t.co/YolKKvXVBg