LinkedIn is better on the app
Don’t have the app? Get it in the Microsoft Store.
Open the app
Discover the latest technologies, resources, events, and announcements to help you build smarter and ship faster. Explore more at developers.google.com
External link for Google for Developers
Google for Developers reposted this
How is redBus building faster, richer UIs? We pulled Amit kumar, their Director of Engineering, aside at #GoogleIOConnect India for a fast-paced chat about all things HTML-in-Canvas. See how this API is leveling up their rich UI experiences and what he's most hyped to build next 🛠️ #BuildwithGemini
Google for Developers reposted this
New aspect ratios. Wider displays. Glanceable interactions. Samsung just unveiled their new devices at Galaxy Unpacked. Ensure your layouts, camera experiences, and widgets adapt seamlessly to these new form factors with our latest guidance and core adaptive design principles → https://goo.gle/4wX99fM
Build web experiences that are optimized for browser and user AI agents → https://goo.gle/4fLmtxU Create reliable developer experiences by exploring: 📦 Tool registration via JavaScript ✍️ Declarative HTML form annotations 🧪 Chrome 149 origin trial
The standard software validation loop takes too much time, time you often don’t have. See how YouTube engineers moved at lightspeed and democratized AI prototyping by rebuilding a full mirrored version of the platform using AI Studio, letting anyone safely test prototypes against live data. 📺 Dive into the full playbook on Emergent: goo.gle/emergent
Google for Developers reposted this
Applications are open for the Google DeepMind Accelerator: AI for the Planet in Asia Pacific! We’re on the lookout for APAC-based startups, research teams, and non-profits using frontier AI to tackle our biggest environmental challenges. Whether your focus is protecting ecosystems, scaling sustainable agriculture, or revolutionizing clean energy, we want to help you take your impact to the next level. 🚀 What’s in store for you? 🤖 Access to Gemini, Gemma, and Google's frontier AI models 🧠 Direct mentorship from Google DeepMind and Google Research experts ☁️ Cloud TPUs and Google Cloud credits to fuel your inference 🌱 3 months of 100% equity-free support 🌍 A global network of founders building a more sustainable planet Ready to build for the future? Apply here: goo.gle/3RsNIUF Know a brilliant founder or researcher who fits the bill? Tag them below! 👇 #AcceleratedWithGoogle
What happens when you deploy generative AI at 100 mph? At Sonoma Raceway, Google Developer Experts built an agentic AI Race Coach grounded in real-time physics and vehicle telemetry. Powered by Google Antigravity and Gemini, the system delivers split-second coaching straight to the cockpit, closing the AI trust gap in high-stakes environments. See how the architecture works and try the Codelab: https://goo.gle/4vHdIcZ
The Build with Gemini XPRIZE competition is officially live ✨ Presented by XPRIZE × Google, this global challenge gives you 90 days to build a real AI business that solves real-world problems. With a $2M total prize pool across 25 winners, over 18,000 builders have already registered. In this 3-part series, learn step-by-step how to leverage Google's ecosystem from raw concept to live production. Watch the rest of the series and learn how to build your startup from the ground up. Start building today → https://lnkd.in/eDSXCeFR Try Gemini → https://gemini.google.com Try NotebookLM → https://lnkd.in/gZaSRHM5 Try Google AI Studio → https://lnkd.in/eBtZYDjq
Training dynamic, tool-using agents exposes a significant infra bottleneck: when your agent pauses to query a search engine or wait on a simulator, your expensive AI accelerators sit idle. To solve this, the latest release of Tunix brings high-throughput Agentic RL natively to the JAX/TPU ecosystem. We've built an LLM-first rollout engine that maximizes efficiency with three core upgrades: 🔄 Async and decoupled rollouts: Overlap model inference with host-side environment execution & streaming trajectories directly to a synchronous trainer via a barrier-free producer-consumer queue to eliminate TPU starvation. 🧩 Composable abstractions: Plug-and-play OSS environments (like SWE-bench) without rewriting your training loops. 🔍 White-Box profiling: Microsecond-level trace integration with Perfetto to see exactly what is starving your hardware. Whether you are reproducing SOTA reasoning models or building custom agents, Tunix provides the high-performance foundation you need. Read the architecture deep dive: https://goo.gle/4px5d2w Explore the code: https://lnkd.in/em3RuPpi #ReinforcementLearning #AgenticRL #Training #MachineLearning #JAX #TPU #OpenSource
Every dev journey needs a solid roadmap. ✨ Whether deepening your knowledge or connecting with other devs, the #GoogleDeveloperProgram Builders Hub is designed just for you. Learn more → https://goo.gle/44BeeOA
