Author Archives:

Build reliable multi-agent applications with ADK Go 2.0. Discover our new graph-based workflow engine, built-in human-in-the-loop, and dynamic orchestration

The Agent Development Kit (ADK) for Go 2.0 has been released, introducing a first-class, graph-based workflow engine to help developers compose complex, multi-agent applications. This update adds built-in primitives for human-in-the-loop (HITL) orchestration, dynamic execution using plain Go code, and automated resilience features like exponential backoff retries. By unifying the execution model, both single-agent applications and intricate graphs now run on the same runtime, simplifying telemetry and state persistence.

Driving the Agent Quality Flywheel from Your Coding Agent

Building AI agents often leaves developers uncertain if prompt tweaks to fix single errors will accidentally cause widespread regressions in production. To bridge this gap, Google has introduced a new developer skill for coding agents that automates a five-stage evaluation flywheel: preparing data, running inference, grading with adaptive AutoRaters, analyzing failure clusters, and executing targeted optimizations. Running continuously against production traffic or on-demand via synthetic scenarios, this tool allows developers to describe testing goals in plain language while an independent evaluation service safely validates and counts actual performance improvements.

AI Overviews in Drive now available on mobile

In April, we announced the general availability for Drive AI Overviews in Drive on the web. We’re now bringing this feature to the Drive Android and iOS apps.

Instead of searching through endless files and opening dozens of tabs to find the information you need, you can now get instant answers right at the top of your search results. Gemini does the heavy lifting for you, scanning your documents to provide clear, reliable summaries.

Here is how it helps you work smarter:

  • See the big picture: Get a quick summary of information pulled from multiple files without needing to open each one.
  • Ask naturally: There’s no need to use complicated search tricks. Just ask a question as you would to a colleague, like "What’s in our Spring 2026 catalog?"
  • Get the right answer: Gemini automatically understands what you’re looking for, whether it’s a quick fact, a project summary, or a list of specific documents, and adjusts its response to match.
  • Dig deeper with ease: If you need more information, you can go from a quick summary to a deeper conversation with Ask Gemini in just one click.
  • Control your AI Overviews scope: Use AI Overview search settings to choose which Google Workspace apps Gemini uses to find files and generate AI Overviews.

This feature will roll out in English and an additional 28 languages (the same as those supposed for Gemini in Drive side panel) over the next several weeks.

Getting started

Rollout pace

Availability

  • Business: Business Standard and Plus
  • Enterprise: Enterprise Standard and Plus
  • Consumer: Google AI Pro and Ultra
  • AI Add-ons: Google AI Pro for Education

Resources

Community feedback: How can corporations improve support for open source maintainers?

We know that AI is actively transforming the sustainability and socio-technical dynamics of OSS communities. Google Open Source is committed to partnering with open source communities and ecosystems to learn together how we should update our own models for engagement and support.

During an open meetup for GitHub Maintainer Month, I led a session to gather community feedback on how corporations can more effectively support open source maintainers.

Paying maintainers takes creativity

Many maintainers would appreciate consistent financial support. However, facilitating payments to individuals without established contractual relationships remains a complex challenge, particularly across diverse international jurisdictions. Fiscal hosts and programs such as Open Collective, GitHub Sponsors, and the LFX Mentorship Program can simplify components of this process, but they do not resolve the underlying issues of funding sustainability and predictability. While initiatives like the Open Source Endowment are working toward long-term funding sustainability, individual maintainers also had a few ideas:

  • Pay per meaningful contribution vs gameable metrics: Avoid payment models based on easily manipulated units like pull request counts or review volume. A proposed alternative is ‘pay per report,' encouraging maintainers to document their achievements and upcoming roadmaps.
  • Commitment-based purchasing: Corporate policies might make procurement simpler (or more complex) than sponsorships, so maintainers could benefit from offering structured support services alongside traditional sponsorship opportunities.
  • Fund conference attendance: In-person networking can be a boon for solo maintainers but it's often cost-prohibitive. For some corporations, travel sponsorship may be a simpler alternative to direct payments.
Challenge for Corporations and Fiscal Hosts: How can we assist maintainers in understanding any and all prerequisites and documentation necessary to participate in monetary programs? Advice from Maintainers: Consult a tax professional to understand the implications of various funding methods.

Manage and respect expectations

Beyond financial support, our discussion returned to the importance of respect and etiquette. Particularly, how can we manage expectations between heterogeneous creators, contributors and users - are maintainers clearly communicating their preferences, and are corporations actively respecting them? Some suggestions include:

  • Adherence to community norms: Maintainers should share their preferred communication channels, while contributors - both human and agentic - must ensure they review and follow them.
  • Consistency with documentation: Discrepancies between documented procedures and actual practices create friction for all participants. This standard should be upheld by both individual maintainers and corporate-managed projects.
  • Clarity of intent: Many maintainers would like to understand the motivation behind a contribution and reserve the right to ask questions.

To improve specific program experiences, maintainers suggested:

  • Consistent communication: Recipients of funding programs expect clearly communicated expectations regarding the timing and amount of disbursements.
  • Transparency and discoverability: Maintainers would appreciate easily discoverable records that track program participation, active agreements, and verify the status of Contributor License Agreements (CLAs).

Let's keep learning as a community

While we cannot make any promises, we want to continue to learn and challenge ourselves to consider novel ways to support OSS communities and maintainers. As a member of our community, we value your opinion. We've created a Google form to collect any thoughts you might have, as well as gauge interest in another open meeting. We plan to share any and all learnings back with the community.

Work with delegated Gmail accounts from mobile devices

Previously, users could only work with delegated Gmail accounts through the web interface. We are updating the Gmail app for iOS and Android to allow delegates to read, manage, and compose emails on behalf of a delegator directly from their mobile devices.

This update removes a significant barrier for employees who rely on mobile devices for their daily productivity. For example, an administrative assistant can seamlessly handle urgent communications for an executive while away from the office, without needing to find a desktop computer.

When using the Gmail mobile app, delegates can now:

  • Switch between their own inbox and delegated accounts.
  • View unread message counts for delegated inboxes from the account menu.
  • See emails intermingled across delegated accounts and their own account using the mobile “All inboxes” view.
  • Send messages that allow recipients to view the specific "sent by" information in the mobile experience. 
Administrators retain full control over delegation settings, including the ability to restrict delegation to specific organizational units. The mobile experience adheres to existing delegation policies and limits, such as supporting up to 1,000 unique delegates per account and 40 concurrent users. Delegators do not need to perform any additional setup to enable mobile access for their existing delegates.

Getting started

  • Admins: There is no admin control for this specific mobile feature; it follows existing delegation settings that you’ve configured for the web experience.
  • End users: To access a delegated account, ensure you have first been granted access via the Gmail web settings. Once granted, tap your profile picture in the Gmail app on Android or iOS and select the delegated account from the list. Visit the Help Center to learn more about delegating and collaborating on email.

Rollout pace

Availability

  • Available in early July to all Google Workspace customers, Workspace Individual subscribers, and users with personal Google accounts

Resources

Ask Gemini in Drive now available on mobile

In April, we announced the general availability of Ask Gemini in Drive on the web. We’re now bringing this feature to the Drive Android and iOS apps.

Ask Gemini in Drive offers you a dedicated, immersive workspace designed for deep focus. You can now engage in high-context, multi-turn conversations to efficiently explore and understand content across Drive, other Workspace apps, and the web.

Key features include:

  • Dedicated conversations: Engage in focused discussions about specific sets of files and folders. By grounding your questions in the relevant content, you get more precise, actionable answers.
  • Persistent conversation history: Easily pick up where you left off. Your past chats are saved, allowing you to quickly revisit previous insights about specific folders or projects without starting over.
  • Secure and compliant: Ask Gemini in Drive is built directly into the Drive architecture, it never copies or replicates your files. It honors your existing data protection and security controls, including access permissions, DLP policies, and IRM, ensuring Gemini only accesses content you are authorized to see.

This feature will roll out in English and an additional 28 languages over the next several weeks.

Getting started

Rollout pace

Availability

  • Business: Business Standard and Plus
  • Enterprise: Enterprise Standard and Plus
  • Consumer: Google AI Pro and Ultra
  • AI Add-ons: Google AI Pro for Education

Resources

Create fully native and editable presentations with Gemini in Google Slides

You can now create a full, multi-slide presentation using Gemini in Google Slides. With a single prompt, you can ground the presentation in existing content from Google Drive, match the style of another presentation, and build fully editable slides, allowing you to make any necessary adjustments. Gemini will also suggest relevant files, emails, and chats that you can choose to add to enrich your presentation.

Try the following to create more relevant, compelling presentations in less time:

  • Add a prompt: In the Slides side panel, add a prompt to generate a presentation.
  • Ground it in your content: Add as many reference files directly from Drive as you need to provide context.
  • Stay on-brand: Attach an existing deck to use as a style reference to ensure your presentation matches your desired look and feel.
  • Refine the plan for your presentation: Answer any follow-up questions to refine the presentation’s tone, style, content, or audience. You will also have the chance to edit or approve the presentation outline before the actual slides are created.

Note: At launch, this feature will be supported in English only.


Getting started

Note: Through at least August 1, 2026, Workspace customers will get promotional access to higher limits for creating multi-slide presentations using Gemini in Google Slides, allowing users to experiment with this feature. Users will see a notification when they use this feature to inform them of the limited higher promotional access period. Per-user usage limits will apply after that date; we’ll provide more information in the Help Center in advance of updated usage limits going into effect.

Rollout pace

Availability

  • Business: Business Standard and Plus
  • Enterprise: Enterprise Standard and Plus
  • Consumer: Google AI Pro and Ultra
  • Education Add-ons: Google AI Pro for Education
  • Other Add-ons: AI Expanded Access*

*Once promotional limits are no longer in effect, users with AI Expanded Access add-on licenses will have higher limits on usage of Gemini in Slides.

Resources