Learn how to build stateful AI agent networks in Python—without vendor lock-in. Save your free spot: goo.gle/agent-builder-ss
Join us live starting September 15 for Startup School: Agent Builder, a 4-part technical series with Google AI engineers.
What we're covering: → ADK & Python: Code custom state transitions with the Agent Development Kit. → Persistent Memory: Connect agents to AlloyDB/pgvector, SessionStore, and MemoryBank. → Production Runtimes: Deploy securely with SPIFFE-based identities on Cloud Run.
Can't make it live? Register to get access to session recordings and technical resources. Register here: goo.gle/agent-builder-ss
🧙♂️➡️Swipe to watch Gemini 3.8 Flash work its magic. It combines native video understanding with advanced coding to autonomously build this 3D game, play it to find bugs, and continuously fix its own code in a seamless agentic loop.
Available now via the Gemini API in Google AI Studio, Android Studio, Google Stitch, and Google Antigravity.
Say hello to Gemini 3.5 Transcribe 🎙️✨ Our newest context-aware speech-to-text model brings deep audio intelligence to streaming and batch workflows. It cleans up filler words, captures complex digits, and uses on-screen context to deeply understand spoken audio.
Swipe through to see 5️⃣ ways to build with it in action.
Gemini 3.5 Transcribe is our latest speech-to-text model built for more precise and intelligent audio understanding and transcription.
Whether you’re building real-time, voice-first interfaces, or transcribing multi-speaker recordings, Gemini 3.5 Transcribe adapts to your needs by:
→ Transcribing complex speech, emails, and phone numbers with lower WER (Word Error Rates) → Adhering to custom vocabulary and specialized jargon → Auto-detecting 85+ languages, regional accents, and live language switches
Start building today! Available in public preview for developers via the Gemini API in Google AI Studio and @googleantigravity , and for businesses via the Gemini Enterprise Agent Platform.
Meet Gemini 3.7 Flash⚡— our most intelligent workhorse model yet for coding and agents. It delivers better intelligence for complex workflows and is available through the end of the year at an introductory price of $0.75/1M input tokens and $3.75/1M output tokens. Now available across Google Antigravity and the Gemini API via Google AI Studio and Android Studio.
We are thrilled to partner with Google Cloud for The Blockbuster Hackathon!
Are you ready to turn real-world enterprise chaos into a seamless cinematic production? Step onto the studio lot for a first-of-its-kind Summer Blockbuster Hackathon with Google Cloud and our outstanding partner ecosystem!
In the era of autonomous builders, writing raw code is just background noise. The real magic is system orchestration.
Choose your role on the backlot:
The Director: Build production-ready autonomous agent networks.
The Technical Producer: Connect secure data pipelines via managed MCP servers.
The Studio Head: Enforce Cloud IAM security and governance across multi-agent workflows.
Vibe code with Gemini Enterprise and IBM tech to cast the perfect enterprise stack and build the next generation of AI applications fit for the big screen.
A dependency resolution dev challenge from a situation every dev has faced at least once. Your app needs three packages. The registry has six versions, each with its own constraints.
The resolver can pick only one version of each package, and it prefers the newest it can get.
Devs, this Kotlin challenge arises from a refactoring bug you might have seen.
Here’s the setup: a teammate tidies up some code that builds job configurations. The build succeeds and everything looks fine. Then we check the list, and instead of the config object we expect, it holds something else.
Google for Developers
Learn how to build stateful AI agent networks in Python—without vendor lock-in. Save your free spot: goo.gle/agent-builder-ss
Join us live starting September 15 for Startup School: Agent Builder, a 4-part technical series with Google AI engineers.
What we're covering:
→ ADK & Python: Code custom state transitions with the Agent Development Kit.
→ Persistent Memory: Connect agents to AlloyDB/pgvector, SessionStore, and MemoryBank.
→ Production Runtimes: Deploy securely with SPIFFE-based identities on Cloud Run.
Can't make it live? Register to get access to session recordings and technical resources. Register here: goo.gle/agent-builder-ss
2 weeks ago | [YT] | 1,259
View 24 replies
Google for Developers
Meet Gemini 3.8 Flash — your newest, sharpest coding partner ⚡
🧙♂️➡️Swipe to watch Gemini 3.8 Flash work its magic. It combines native video understanding with advanced coding to autonomously build this 3D game, play it to find bugs, and continuously fix its own code in a seamless agentic loop.
Available now via the Gemini API in Google AI Studio, Android Studio, Google Stitch, and Google Antigravity.
Start building today: goo.gle/46B6Fs3
2 weeks ago | [YT] | 898
View 49 replies
Google for Developers
Say hello to Gemini 3.5 Transcribe 🎙️✨ Our newest context-aware speech-to-text model brings deep audio intelligence to streaming and batch workflows. It cleans up filler words, captures complex digits, and uses on-screen context to deeply understand spoken audio.
Swipe through to see 5️⃣ ways to build with it in action.
Read the blog to learn more: goo.gle/4qMie9m
3 weeks ago (edited) | [YT] | 611
View 14 replies
Google for Developers
Gemini 3.5 Transcribe is our latest speech-to-text model built for more precise and intelligent audio understanding and transcription.
Whether you’re building real-time, voice-first interfaces, or transcribing multi-speaker recordings, Gemini 3.5 Transcribe adapts to your needs by:
→ Transcribing complex speech, emails, and phone numbers with lower WER (Word Error Rates)
→ Adhering to custom vocabulary and specialized jargon
→ Auto-detecting 85+ languages, regional accents, and live language switches
Start building today! Available in public preview for developers via the Gemini API in Google AI Studio and @googleantigravity , and for businesses via the Gemini Enterprise Agent Platform.
Read the blog to learn more: goo.gle/4gUy2TW
3 weeks ago | [YT] | 999
View 32 replies
Google for Developers
Syncing to and from @GitHub is now supported in ai studio build.
Start from an existing repo, push and pull changes, and seamlessly work across environments
Get building → goo.gle/4xfRFf7
1 month ago (edited) | [YT] | 594
View 22 replies
Google for Developers
Meet Gemini 3.7 Flash⚡— our most intelligent workhorse model yet for coding and agents. It delivers better intelligence for complex workflows and is available through the end of the year at an introductory price of $0.75/1M input tokens and $3.75/1M output tokens.
Now available across Google Antigravity and the Gemini API via Google AI Studio and Android Studio.
Learn more → goo.gle/4znXpVF
1 month ago | [YT] | 1,196
View 66 replies
Google for Developers
A dependency resolution puzzle about a failure mode that hides from normal debugging.
Three packages share a single dependency, and each declares its own version constraint. Checked in pairs, every combination is compatible.
The question is what happens when all three are installed together. → youtube.com/shorts/qwUM4Sk1_nE
Watch the clip and comment: which version of the shared dependency gets installed?
1 month ago | [YT] | 330
View 19 replies
Google for Developers
We are thrilled to partner with Google Cloud for The Blockbuster Hackathon!
Are you ready to turn real-world enterprise chaos into a seamless cinematic production? Step onto the studio lot for a first-of-its-kind Summer Blockbuster Hackathon with Google Cloud and our outstanding partner ecosystem!
In the era of autonomous builders, writing raw code is just background noise. The real magic is system orchestration.
Choose your role on the backlot:
The Director: Build production-ready autonomous agent networks.
The Technical Producer: Connect secure data pipelines via managed MCP servers.
The Studio Head: Enforce Cloud IAM security and governance across multi-agent workflows.
Vibe code with Gemini Enterprise and IBM tech to cast the perfect enterprise stack and build the next generation of AI applications fit for the big screen.
Learn more → goo.gle/4fr50uw
1 month ago | [YT] | 563
View 10 replies
Google for Developers
A dependency resolution dev challenge from a situation every dev has faced at least once. Your app needs three packages. The registry has six versions, each with its own constraints.
The resolver can pick only one version of each package, and it prefers the newest it can get.
Which versions get installed if any? → youtube.com/shorts/UUozyqDNRGg
1 month ago | [YT] | 272
View 18 replies
Google for Developers
Devs, this Kotlin challenge arises from a refactoring bug you might have seen.
Here’s the setup: a teammate tidies up some code that builds job configurations. The build succeeds and everything looks fine. Then we check the list, and instead of the config object we expect, it holds something else.
Watch the video and share what you think caused the bug! → youtube.com/shorts/AaWdVp_py4Y
2 months ago | [YT] | 301
View 16 replies
Load more