Learn how chex.checks.chexify runs value-level assertions inside jitted JAX functions to catch NaNs, detect unexpected JITs, and debug Flax NNX models—plus tradeoffs.

Catch Value-Level Bugs in JAX: Advanced Chexify for Flax NNX

Enhancing Reliability (Part 2): Advanced Chex Techniques for JAX and Flax NNX

This second episode of our two part series on Chex builds on the fundamentals, introducing advanced tools for debugging values and performance, and then shows how to apply everything within a Flax NNX model.  

Resources: 
Learn more → https://goo.gle/learning-jax

Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Robert Crowe

Enhancing Reliability (Part 2)

A practical guide to the JAX AI stack: XLA-powered jit, grad and vmap plus Flax, Optax, Orbax and Grain for simple, fast, scalable ML development.

Practical JAX AI Stack Guide — jit, grad, vmap & Flax/Optax

Leveraging the JAX AI Stack

In this video we present an overview of the JAX AI Stack, focusing on how it can be used for high-performance model development and training. We'll draw parallels to PyTorch concepts, which you may be familiar with, to highlight the similarities and differences.

Resources: 
Learn more → https://goo.gle/learning-jax

Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Robert Crowe

See how AI-First Colab and Gemini turn raw data into interactive tables, time series, and animated maps—no plotting boilerplate. Workflow, examples, and export tips.

Create Interactive Data Visuals Fast with AI-First Colab & Gemini

Stunning Visualizations with AI-First Colab

Ready to visualize your data without writing a single line of code yourself? This video on visualization with AI-First Colab demonstrates how your Gemini-powered coding partner instantly turns your natural language requests into sophisticated, interactive data visualizations. You'll see how to generate a variety of powerful data outputs, including an interactive table of weather station data, a time series plot comparing temperatures in major cities, and an incredible animated heat map showing temperature movements across the world. AI-First Colab handles all the complexity, allowing you to focus on interpreting your results and telling your data story. See how you can accelerate your visualization workflow today.

Subscribe to Google for Developers → https://goo.gle/developers 

Products Mentioned:  Google AI, Colab

Technical deep-dive on securing autonomous AI agents: learn about prompt injection, data leakage, excessive permissions, and a 'Model Armor' pattern for layered defense.

Defense-in-Depth for AI Agents: Stop Prompt Injection & Data Leaks

How to Secure Your AI Agents: A Technical Deep-dive

AI agents introduce unique security challenges like prompt injection, data leakage, and excessive agency. This workshop breaks down the OWASP LLM Top 10 and demonstrates how to implement robust security controls using Model Armor and the Agent Development Kit (ADK). Learn how to filter malicious inputs, redact sensitive data from outputs, and manage authentication patterns to build secure and reliable agentic workflows.

Resources: 
Explore the Agent Development Kit → https://google.github.io/adk-docs/
Subscribe to Google for Developers → https://goo.gle/developers

Chapters:
0:00 - Introduction to agent security
1:32 - OWASP LLM top 10 vulnerabilities
2:56 - Sensitive information disclosure
4:10 - Excessive agency risks
5:11 - Input filtering with Model Armor
6:26 - Demo: blocking prompt injection
10:18 - Output filtering and PII redaction
11:09 - Demo: redacting sensitive data
12:43 - Authentication and authorization patterns
13:44 - Agent-to-tool authentication flow
16:13 - Managing API keys with Secret Manager
18:05 - Security for A2A and MCP protocols
19:22 -  Supply chain security measures
22:25 -Protecting agent access and integrity 

Subscribe to Google for Developers → https://goo.gle/developers  

Speaker: Sita Lakshmi 
Products Mentioned:  Agent Development Kit (ADK) 

#AIAgents #ADK #AgenticSecurity

How to secure your AI Agents: A Technical Deep-dive

Recap of GDG Summit MENA-T in Dubai: 130 organizers from 72 chapters aligned on practical AI, Cloud, and community playbooks — workshops, governance, and templates to scale chapters.

GDG Summit MENA-T Recap: Practical AI, Cloud & Community Strategies from Dubai

GDG Summit MENA-T: How 130 Organizers Aligned on AI, Cloud, and Community in Dubai

The MENA GDG Summit 2024 brought together 130 Google Developer Group (GDG) organizers from 72 chapters across 20 countries in Dubai to align on Google's strategy, especially around AI and Cloud. Watch now for highlights from the 3 parallel tracks of sessions and workshops!

Top Googlers, including Saad Ouchkir (Director of Customer Engineering, MENA) and Vikas Anand (Director of Product Management), were on the ground, ensuring maximum support and cutting-edge content.

Subscribe to Google for Developers → https://goo.gle/developers 

Products Mentioned: Google Developer Groups

GDG Summit MENA-T

Nikita Namjoshi explains how LLMs 'think': scaling laws, chain-of-thought prompting, test-time compute, best-of-n and reward models — practical tips for developers and curious users.

Inside Thinking LLMs: Scaling, Chain-of-Thought & Test‑Time Compute

How do thinking and reasoning models work?

LLMs that can "think" and "reason" have become increasingly popular. But what is a model actually doing when it's "thinking" and how can we train LLMs to be better at reasoning? This explainer video covers the fundamentals of how thinking models work, including concepts like scaling laws, test-time compute, and reinforcement learning from verifiable rewards.

Resources: 

Gemini thinking → https://goo.gle/3KaUL0J 

Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Nikita Namjoshi 
Products Mentioned:  Google AI

Turn painful framework upgrades into a single automated workflow. Learn how Google Antigravity plans, executes, and verifies migrations — with a concrete Flutter/Dart example.

Automate Framework Migrations with Google Antigravity

Simplify Framework Updates with Google Antigravity

Discover how Google Antigravity, an agentic development platform, simplifies complex framework updates. Learn how to update a Flutter and Dart application to its latest version, eliminating manual migration headaches. Moreover, see Antigravity's autonomous agents plan, execute, and verify complex tasks, including parsing change logs, fixing deprecation warnings, and ensuring application stability. See how you can improve your development workflow and accelerate building processes.

Resources: 
Download Google Antigravity → https://goo.gle/AGY 

Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Emma Twersky 
Products Mentioned:  Google AI, Google Antigravity

Simplify framework updates with Google Antigravity

Bite-sized Canva hacks to speed your workflow: cluster ideas with Whiteboards Sort, create editorial product shots with Background Generator, and use live presentation shortcuts.

Canva Tips & Shortcuts: Faster Brainstorms, Editorial Product Photos, Presentation Tricks

Quick design tips with the Canva team ✏️

Get ready to boost your creativity and productivity  ✨ Join a few members from the Canva team as they reveal their top tips and insider knowledge. Whether you're just starting out or already designing daily, you’ll walk away with ideas you’ll want to try right away!

🔖 For more lessons, tutorials, and certifications, go to: https://www.canva.com/design-school
💬 Any questions? Leave a comment below, and our team will get back to you.

#ProductivityTips #CanvaTutorial #CreativityTips

DeepL explains Arion — a DGX SuperPOD AI platform using TensorRT‑LLM, NVFP4 quantization and efficient models to deliver fluent, near‑zero latency translations at global scale.

Inside Arion: DeepL’s Low-Latency AI Infrastructure for Real-Time Translation

Inference at Scale: How DeepL Built an AI Infrastructure for Real-Time Language AI

DeepL, a global AI product and research company focused on building secure, intelligent solutions to complex business problems, has achieved unprecedented speed and accuracy in AI translation by deploying context-dependent AI translation in milliseconds with NVIDIA and EcoDataCenter. 

The company's AI infrastructure, named “Arion”—powered by an NVIDIA DGX SuperPOD with DGX GB200 systems—enables millions of daily users across dozens of languages to receive accurate translations with near-zero latency. 

This video highlights the collaboration with DeepL, NVIDIA, and EcoDataCenter to provide highly accurate, fluent, and context-aware AI language translations that feel natural to native speakers. 

In addition, DeepL is leveraging NVIDIA TensorRT LLM and NVFP4 inference on NVIDIA GB200 NVL72 systems to train Mixture of Experts (MoE) models, advancing its model architecture to improve efficiency during training and inference, setting new benchmarks for performance in AI.

Build AI fluency with the 4D framework—Discover, Design, Develop, Deploy—to help nonprofits automate routine work, improve reporting, and scale donor outreach responsibly.

AI Fluency for Nonprofits: A Practical 4D Guide to Responsible Use

AI Fluency for Nonprofits: A Practical Guide to Confident, Intentional AI Use

A trailer of AI Fluency for nonprofits developed by Anthropic and Giving Tuesday. 

View the full free course, including all videos, exercises, and resources, at https://www.anthropic.com/ai-fluency-for-nonprofits

This video is copyright 2025 Anthropic PBC and Giving Tuesday. Based on the AI Fluency Framework developed by Prof. Rick Dakan (Ringling College of Art and Design) and Prof. Joseph Feller (University College Cork). Released under the CC BY-NC-SA 4.0 license.

AI Fluency for nonprofits course trailer

GPT-5.1 makes reasoning the default—fixing context and tone issues, improving instruction-following, and using model switching plus personality controls for smarter, more human-friendly responses.

GPT-5.1: Why Reasoning-First Design and Model Switching Matter

Shaping Model Behavior in GPT-5.1 — What Changed and Why It Matters

What does it mean for an AI model to have "personality"? 

Researcher Christina Kim and product manager Laurentia Romaniuk talk about how OpenAI set out to build a model that delivers on both IQ and EQ, while giving people more flexibility in how ChatGPT responds. They break down what goes into model behavior and why it's an important, but still imperfect blend of art and science.

Chapters
- 00:00:43 — GPT-5.1 goals and the shift to reasoning models
- 00:02:18 — Differences between GPT-5 and GPT-5.1
- 00:04:55 — Unpacking the model switcher
- 00:07:24 — Understanding user feedback
- 00:08:27 — Measuring progress on emotional intelligence
- 00:10:02 — What is model personality?
- 00:14:25 — Model steerability, bias, and uncertainty 
- 00:21:59 — Advantages of memory in ChatGPT
- 00:25:27 — Looking ahead and advice for getting the most out of models

Shaping Model Behavior in GPT-5.1— the OpenAI Podcast Ep. 11

Use Claude.ai like a collaborative teammate: learn a three-part prompting formula, how to upload files, connect tools, choose models, and run Research mode for deep analysis.

Claude.ai Playbook — Prompting, Files, Models & Research Mode

Getting started with Claude.ai

Learn how to get the most out of chatting with Claude. This tutorial covers the basics of Claude's conversational interface, including how to craft effective prompts, upload supporting documents, use search and tools, customize your experience with styles and model selection, and leverage features like extended thinking and research mode. Whether you're new to Claude or looking to level up your skills, this video will help you work more effectively with your AI collaborator.

Archive