ase  
Studio<https://notifications.google.com/g/p/ANiao5pzUN_YqQHMkCOB2ljE-FY-R8msfDBWqILYQjZNpicVXtSzQ8sQOn7wanHh9rf0OPIAvwdYpnA8vnzP2rdCJTrXvbLcdbFNiS3F5LcAsufIaJQW9-x_u-TMUc8uibZcT0uuofWOw0Ti0tU45q4pm5wwoXpJ09HjrHe0FEhcuB3NUvNJEIKsCfCYCkJwcNRjepjFiTg5WBXZDPrGsNSrfMsyRK_Xv_VvH3qZ1hNKrt9cMXuYOvdwpIYNSu7zeER7Y6fEH4qyZKOnBXJqBqwurssHtPtojI_xvIOo8z68_3Y>,  
an exciting new experience for Firebase developers. As a member of the  
Google Developer Program, take advantage of 10 Firebase Studio Workspaces  
as part of your membership today. Even better, if you upgrade to Google  
Developer Program  
premium<https://notifications.google.com/g/p/ANiao5rbKSVfyGbRajjR_ksdK9GmDwytKopEwjrEYS1gEhYSip2_ryrqoHSG9RTI4KQ8rhi7AcitOU0OZXIMYH3PbDHz2nenncY0anULsam5K4EYhkGXjFUjzMgBTMVeCvOUHwA3gUaT4NrQfkamJKtbmf9oijtQKvk>,  
you'll have access to 30 Firebase Studio Workspaces for even greater  
flexibility to handle complex projects. The text-to-app feature is  
excellent, and everyone should take it for a spin.

Watch this

Let’s learn about AI agents.

Jason and Aja are back with another season of teaching AI concepts to us  
mere mortals. This video helps us understand AI agents versus AI workflows.

https://notifications.google.com/g/p/ANiao5rfinaF6PeoGSH-D5YHAbDGNeDq3dnpZvjRDv0y705HNv9GQVX043uGxIB77cJS0oVAh6Mcvk1_ynoVUKQUPkZcTPQmVfYSCwa7xE_WXysfTT43ZnUDT54saC8qZZ9Kpt0YuYPTA44cNO3xAsJZgWRA2-dynD-YifmXdoZVRfWO

Community cuts

Let’s take a deep dive into Model Context Protocol.

I liked this post from  
Arjun<https://notifications.google.com/g/p/ANiao5qlcRoBIWyXOz06cxw7pNMleVUbAs1Z9F7yWYW8U4ffKcWZzip1d8BPwLRIqu8mkk-nuOsyGcDfB1V8PhOe-7Sna4ZRL__Hx1rvn79DN3VgdVilb6aC-6SH02g4Rn98iSaiJtAQt6yph5jC5i8xxu7IZLWGP9bP4W90e_PreCbH7_3fkTfTcqq7tVVQ4Z_2KjKPtymuu8_lcPxQW92GoIksu9v3cAJcJk-OCztEG5K80xrtq3IEidVsZ0aZB-GGvoX17jZ4kq4idgMuacq6wfk>  
that explains what MCP is all about and how to use it with Gemini. Lots of  
code to follow along with!

Set up GKE for weighted load balancing.

This feature snuck out without much fanfare, so I’m happy to see this blog  
post<https://notifications.google.com/g/p/ANiao5qS5x3SshCmHpgE5vF-iXQU9LVfYxU1iyuYNzvHT-fM-etTV6T4axmMxZgbf9HRrJXElVwRqtVI0XqeA-D7_mpozO5E2E59Hm-8jjoBglubnyG88ZntRYjSNYDXR6zd55-sL0K8fqR_s-Pkxjfl9th4Jb4zTtFFPxmeOFF_iC5Q78hcqxtILIGf-_mCGR7BJOA1BdGkbBVsleYjHCk9MnCT1aZTYIeHOXVNuPQ>  
which explains this new feature of Google Cloud load balancing.

Apply structured output to your Gemini responses.

You don’t want to be parsing unexpected response types from your LLM.  
Saverio  
explains<https://notifications.google.com/g/p/ANiao5rBIz5wHlUhz-Tw7WCDVHr_WIFG8GmONlUpajbuWGTkAqmHjED_dcgpZRApBRgPT95lGw7qDuiVDNBkQv1_DCKm9PHkhi0quIaNVfLAdzzZ3WPOLa0rYRlkWc40plprJVbeV5banHERemi5orK2Ei5cXcajTTuf-lfOrm-fkS4cRVHrgtwmCGVN-uUB6XQGtcKAFOiCbe-3d9rcvenMppmQKbkKid596YgfdMHW6VcEEQ1p69TI7Gga_gtymWcQ15bjrWtmV4fbgYo>  
how to direct Gemini to put responses into a user-provided schema that you  
can rely on.

Learn and grow

Host an LLM as a sidecar for your serverless app? Sure.

This is such a wild pattern and I love it. Walk through this  
codelab<https://notifications.google.com/g/p/ANiao5pjoWkhhDD1R3UhNA_bEY3Waz1h2UfrbTXedfkhn-xKUbQVS_KqtYY8HJEz7N8-6yjRD_0_o5tQBUO9vQ0wSm7U4nzbmuxZSG9tIZo4A-jQHzpmimUF6KRd4YQWO7e_ywIvGJGCu3bXLxCT_bNidxtjZLVZFboHBm45HoBWMBgDMolabnpU2u4fH-kF1uHCSIPnfQfuLz1Kug0XSkxnDrrw9iOF3D3sQZD9_jtdlhtd2Wey7hGfTPqcKAyf-MOMFVaczx6OO7GDBcaBinoq0vKb>  
to see how (and why!) you’d run Gemma as a sidecar for your web app.

Benchmarking GKE at 65,000 nodes.

What sort of tests, tools, and metrics might one use to run a workload test  
on a massive Kubernetes cluster? You’ll have all the answers after reading  
this.<https://notifications.google.com/g/p/ANiao5opSEAjwbGJjNdm4ajnwYFBr4yI8O-fNpE77cEruTF7y6LKA2AToy_7jhR5q6iQt1c8F2lG6H-07RDLVpPPtuMrlZaf59sZnXXOn5IXfDwW8CEOudcuubV42bNxILcstrRIFNBYCh4GScbBGN-DCRxCUk1QsfmDiSpvo-HVA6eCe931SEBBLzm5ctfqIyUbFin-6rEU1G-1FbCXpMcC60BcpTwASJOoiW8HRXe5zINpxKtcIxUZD4zaOds2aaQitTaRIgEYr_VO7JLf5Ug_-C2K8Yynn0S9Vm-RneiutSU>

Need help going from earlier Gemini versions to the latest? Read this.

It’s time to plan the move from PaLM and Gemini 1.x models to the 2.x  
family. These  
docs<https://notifications.google.com/g/p/ANiao5pFRljayGYVSKl9FYTKPY2j1qdobH_KmiinPk71sum1Tck9EM1gWyhX6elt0-Fv8fCexLU0jSJZNUKh-5cfm0hCN2bqHzqF828cvctyBjtwGXVq_gseDjVYnseoQs8Tmv2oWAUYFwx8u-ALhLEgcmm3CYzEaCygF9sL03leJrtQtpQ_PQS8XsbZw9h3rUUopIJUXN-qqw>  
offer some helpful guidance.

One more thing

Video generation through the Gemini API.

Logan shared that you now have access to Veo 2for text to video and image  
to video scenarios.

https://notifications.google.com/g/p/ANiao5pHlRJlRkFMV3zrzy_NA1yk4O42rZW5IMlKDUa3Z-LKcfciYtGqDwHXTATX3mxaubKDaVsxTYO_7iuSr79eFSbf_O-_GsGPGnOxyBdkvGEXAdXtXJioqpKUVo0eHMJeqah5oJ3s24qHfv79V-HqqTsTEjUZy4so0E4w60J5BwFAwBHwDrP7ur5ZgignCgh4VsA8

Follow us on

Social media.

Add your voice to the conversation and
get the latest news, views, and insights.

x<https://notifications.google.com/g/p/ANiao5ocsKnn1ywuYF7grzPTqnnyKAligAkslcMsip47bvFvQ9CVMUspAKFNMzINnpZ080estlGbY3CwGVf4sX2KQAmPyjycNaIBcImo0mJvbQMMQC7sTXDSgA>

Google Cloud  
blog<https://notifications.google.com/g/p/ANiao5pg6xSiJMRmTjOXkGuq8tHvFjqRUcvrZZrMwCDErsRg8YP-ZeBXUO3pDkGvjPbyEOZntlX-HVK47zEezRXBYEnMzm-R47kg0K5Ap7k-fbiMQQWh5fnlG6CsnR_UVu4H>

LinkedIn<https://notifications.google.com/g/p/ANiao5oFbLRdvCAoK-P0XYRYocChE4cRDTP0sNibKV6FbH4MOiH6qHzn0DPMx9Sq9HXVyqw9LoXdeu_InZZ31lXvFCKYHs5w1vOjzSqvtvUJxecXsS3V-wdX1-3Wf6zLXGZbjI26aAtxjT1n_kUyC5Drjh8gytmu>

Facebook<https://notifications.google.com/g/p/ANiao5rbD8H_WJz0HW39hOB1VBwPzGfY6_gdUXXMuslyLgpTU3stOBeQf5nAweofORNMuaq8mDNncOl2ipLlKXOPMJwjv3THsYYeYncYKXcX8i9TDSwWAGP3VxxZIk5K-31gFxLz4DyQc9yNdok>

YouTube<https://notifications.google.com/g/p/ANiao5rA6hSx9VhtOmrhhHueEmdTvajosrh-Z5zxFtDGo9FWiwy0aMaKM81aGSRn1xYRl1RmroUnEuqWjT0nuYSqp_n7mpPgu_ImpxafIkmAmRE0Cs06nHYu3KKB_Zzz7SeYK00TG9UAsvGjNwlX_P9m5g>

GitHub<https://notifications.google.com/g/p/ANiao5pmINX-SEVczxIPk0_j8c_gs29bdOR7gh1ONmmTKY5p5Bq1K45ayxwDDMozqpql2sfHFMMAHdtshTlHOicYENDtQa-DxaIl3ZzmvRsjxv1_QDU-Ka2fFBoxdRuzA7I6wN2mg192JOZyPw>

No-cost cloud training

35 credits on Google Cloud Skills Boost. Redeem  
now<https://notifications.google.com/g/p/ANiao5oOJ43GqACJcobT5zDgzH0--TmBdIoyWMkU89Kts7k3Z8Vbm4ysp0DXIoxibqFin0OCb3iZvgbBaE-yjNZl0dh5g2NjOT6zPBT3ztgPaQeeyQ5uo7cQ-DRlYydYRJn9fogIxy9Utcx9OHFPc9zR2npk2edHfLykAN-MnpeDFXbDTsfdW9cp16Mfwgf8AvTQTN8fCggAyMZmliudP9bUg1CQhR8w0oE_FW8hVDrbBJTybVfEKB-0GOGuoaaeaeYZRyRSFMUxLaL6HwNpeIln5tccj7n5B7TccjhvTdr5rjPV-_5arz04deH8HzQu_iLX1dd6pCpN360FTYxp6YEFzunb-dHUYpBcKRy_MTO3QRbgo27j>
Already on Google Cloud? Go to your  
console<https://notifications.google.com/g/p/ANiao5o7y8K6T80HGBtniEX9ZJrM91MpTZIzYxT3kzK5h1y9wN8UGMlWQnBDrv8ne9e7qK6QmiGIM7UBQuo0IBw04fHolH3l3psV1Vjy58FitNCbHir_hxlgyxP8wzu5dULbVnAXx9uhg-QT62mmpcc>

Google LLC 1600 Amphitheatre Parkway, Mountain View, CA 94043

This email was sent to pfeifferj@archlinux.ch because you indicated that  
you want news and program announcements from the Google Cloud Innovators  
program. If you don't want to receive such emails in the future, please  
unsubscribe.  
<https://myaccount.google.com/communication-preferences/unsubscribe/gt/ANiao5odktOssMQP0ywAspAcYWDtQ6nEwVHH5UXTbh0ks-_pG3dcMw0J8uIfnW-Bdtb9k095QJeNIqDsRJyDbdPamxyG0yvPdrKuYdk3WPnrwBQu2l6go3sTb6bG4CMdQl7xK9jJAgj2wQOcd8h2r7u3SnnkLnprWgtgYlDgpzOXtkR98jSQ6PYotgj6e42kl4Uf_PCXEOFgpmabQttwLmADv5lrqsqzq8SyzOfZnc8-ibZjbxv4f1Sc0P5z6iUuxiEx4qReyHZG272wUnm0A-kafqfFkWBePwng?utm_source=gm&utm_medium=email&auto=true>  
You can also change your preferences by visiting  
https://developers.google.com/profile/u/me/my-community  
<https://notifications.google.com/g/p/ANiao5p71OTuzZQReQkKkU3vEUEPoaiFc1ausmUJUMYoLY9rYbuaWzumK-6QrvEyjGUYoNG8_m07p7jk6c44wHfiYu1Rr-kowtMjCBKhWD6iNOTs8QFdcuxOZubg9twrNGh4sFtUEojyfosamcx5O2C0vxfIaQDee2haTULScipryog>  
and editing your selection under “community email preferences” for Google  
Cloud Innovators.The overwhelmed person’s guide to Google Cloud: April 11Google Cloud Innovators <google-cloud-innovators-noreply@google.com> undefinedpfeifferj@archlinux.ch pfeifferj@archlinux.ch