Agents are becoming part of a massive, interconnected ecosystem. They increasingly rely on tools, skills, and other agents distributed across teams, organizations, and platforms. For this ecosystem to scale, agents need reliable answers to three questions: Where does the right capability live? Which capability should I actually use? How do I verify it’s safe to connect to? Today, there is no standard way to answer those questions across organizations. That’s why, alongside industry partners including Cisco, Databricks, GitHub, GoDaddy, Hugging Face, Microsoft, NVIDIA, Salesforce, ServiceNow, and Snowflake, we’re proud to announce the Agentic Resource Discovery (ARD) specification. Read the full announcement: https://goo.gle/4a2sTWf
Google for Developers
Technology, Information and Internet
Mountain View, CA 4,111,049 followers
Join a community of creative developers and learn how to use the latest in technology—from AI and cloud, to mobile & web
About us
Discover the latest technologies, resources, events, and announcements to help you build smarter and ship faster. Explore more at developers.google.com
- Website
-
http://developers.google.com
External link for Google for Developers
- Industry
- Technology, Information and Internet
- Company size
- 10,001+ employees
- Headquarters
- Mountain View, CA
- Specialties
- coding, engineering, firebase, android, cloud, web development, and mobile development
Updates
-
Google for Developers reposted this
Chrome DevTools for agents changes how AI optimizes your web apps. Instead of letting a coding agent guess from static source code and burning tokens, this allows your AI to run actual performance traces and agentic web Lighthouse audits in a real browser. #GoogleIO
-
How do we unlock the true power of AI in development? When we integrate AI into software and product development, it’s natural to start with code generation. It’s an incredible tool, but it's just the beginning of what's possible. To experience the full impact of AI, we have the opportunity to expand our vision in two exciting ways: 1️⃣ Embrace the entire lifecycle: The real magic happens when we welcome AI across the entire software and product development journey - from early planning and design, through building and testing, to deployment and operations. 2️⃣ Evolve our processes: The most rewarding transformations happen when we use this new technology to help our traditional workflows naturally evolve, opening doors to smarter, more collaborative ways of working. AI isn't just about doing the same things faster - it’s about discovering entirely new horizons for innovation. Ready to dive deeper into the future of product development? Experience the full session here: https://goo.gle/3Q7eWj7
-
Improve your gen AI developer workflow with the latest Google Gemma updates ✨ → https://lnkd.in/e-jqJzik Luciano Martins and Gus Martins from Google DeepMind are hosting a live AMA session entirely in Portuguese. Join the discussion live on stage this Friday, June 19 to discover fresh insights, optimize your builds, and get answers directly from the team working on the Gemma open models family.
-
-
Google for Developers reposted this
This past weekend, I had the honor of collaborating with an incredibly smart and creative group of people on a very complex challenge: building a trustworthy, on-device AI coaching system capable of interpreting high-speed racing data and delivering actionable feedback inside performance vehicles running at triple-digit speeds. We moved beyond a prototype by embedding an agentic AI stack directly into high-performance vehicles under real track conditions. A special shoutout to Brian Luc and Angelo for their exceptional systems integration work, which helped connect our AI stack with high-quality telemetry data. Across the cohort, we had three teams led by Ajeet Mirwani, three cars, different sensor configurations, and three unique application approaches. For our system, we structured the architecture around three main components: Vikram Tiwari Sebastian 𝟭. 𝗠𝗲𝗺𝗼𝗿𝘆 𝗕𝗮𝗻𝗸 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗼𝗿 A web platform where coaches and drivers can evaluate, visualize, and discuss previous performance sessions. With the power of Gemini, the platform transforms multimodal datasets, including telemetry data, screenshots, plots, and performance snapshots, into AI-generated coaching notes. These notes can be exported as a form of “brain-transfer learning” and imported into the on-device app, where Gemma uses them as contextual memory for real-time coaching. 𝟮. 𝗗𝗮𝘁𝗮 𝗩𝗶𝘀𝘂𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗗𝗮𝘀𝗵𝗯𝗼𝗮𝗿𝗱 A trackside data visualization dashboard that allowed us to analyze logged sensor data extracted from the on-device app after each track session, identify bottlenecks, and debug overall system performance. The dashboard also included a built-in simulator, enabling us to test and refine the mobile app outside the vehicle environment. 𝟯. 𝗠𝗼𝗯𝗶𝗹𝗲 𝗔𝗜 𝗖𝗼𝗮𝗰𝗵𝗶𝗻𝗴 𝗔𝗽𝗽 The most critical component was the mobile AI coaching app. It provided two types of coaches: a deterministic, heuristic physics-based coach and a Gemma-based coach powered by an optimized on-device version of Gemma, supported by contextual memory generated from previous sessions. To improve latency, we tested Gemma across CPU and GPU, explored speculative decoding through Multi-Token Prediction, and evaluated different on-device runtimes, including MediaPipe, ML Kit, and LiteRT. The goal was to deliver concise, timely, and actionable voice coaching to the driver while on track. Special shoutout to our pro driver, T-Rod, who broke his personal record while testing the app and gave us the perfect research question: how useful is an AI assistant when the human is already driving at that level? Jokes aside, it made the test even more valuable. Google Antigravity played a key role in accelerating our development process, helping us debug, reason through system issues, and move faster as a team. Huge respect to every collaborator in this cohort. Chantelle Taha Rabimba Adrian Hemanth Aileen Simon Francisco Ana Jorge Google Developer Experts Google Google for Developers
-
-
-
-
-
+6
-
-
From live talks to international meetings, voice translation just got seamless. More natural-sounding speech, no awkward pauses. Now available in preview, watch how Gemini 3.5 Live Translate handles speech-to-speech translation for global audiences. Developers can build experiences where attendees join a session and listen along in their preferred language. The model even recognizes language switches automatically when the speaker shifts languages! Try it in Google AI Studio: https://goo.gle/4oluWL5 Get demo code: https://goo.gle/4oin2BS
