Multi-agent workflows often fail. Here’s how to engineer ones that don’t.
Most multi-agent workflow failures come down to missing structure, not model capability. Learn the three engineering patterns that make agent systems reliable.
Building Semantic Search on my Content
I've added some pretty cool AI-powered features to kentcdodds.com and I want to tell you all about it.
Helping YOU ask ME questions with AI
Here's how I've made it easier for you to call into the Call Kent podcast without having to record yourself and also make yourself anonymous using AI.
RCCLX: Innovating GPU Communications on AMD Platforms
We are open-sourcing the initial version of RCCLX – an enhanced version of RCCL that we developed and tested on Meta’s internal workloads. RCCLX is fully integrated with Torchcomms and aims to empower researchers and developers to accelerate innovation, regardless of their chosen backend. Communication patterns for AI models are constantly evolving, as are hardware […]
How AI is reshaping developer choice (and Octoverse data proves it)
AI is rewiring developer preferences through convenience loops. Octoverse 2025 reveals how AI compatibility is becoming the new standard for technology choice.
How Stably ships AI testing agents in hours, not weeks
Automate repository tasks with GitHub Agentic Workflows
Discover GitHub Agentic Workflows, now in technical preview. Build automations using coding agents in GitHub Actions to handle triage, documentation, code quality, and more.
How low-bit inference enables efficient AI
Making products like Dropbox Dash accessible to individuals and businesses means tackling new challenges around efficiency and resource use.
Designing For Agentic AI: Practical UX Patterns For Control, Consent, And Accountability — Smashing Magazine
Autonomy is an output of a technical system. Trustworthiness is an output of a design process. Here are concrete design patterns, operational frameworks, and organizational practices for building agentic systems that are not only powerful but also transparent, controllable, and trustworthy.
Why demand for designers is on the rise
Our latest study suggests that AI is driving renewed momentum in design hiring. We unpack why that is, what hiring managers prioritize, and which skills designers need to get ahead.
How we built AEO tracking for coding agents
Learn how we built an AI Engine Optimization system to track coding agents using Vercel Sandbox, AI Gateway, and Workflows for isolated execution.
Anyone can build agents, but it takes a platform to run them
Why competitive advantage in AI comes from the platform you deploy agents on, not the agents themselves.
Continuous AI in practice: What developers can automate today with agentic CI
Think of Continuous AI as background agents that operate in your repository for tasks that require reasoning.
The Vercel AI Accelerator is back with $6m in credits
A six-week program to help you scale your AI company offering over $6M in credits from Vercel, v0, AWS, and leading AI platforms
For the love of craft: Vectorize images in Figma
With our new AI image editing tool, Vectorize, you can turn raster images into editable vectors, allowing you to tweak, refine, and scale designs directly in Figma.
Pick your agent: Use Claude and Codex on Agent HQ
Claude by Anthropic and OpenAI Codex are now available in public preview on GitHub and VS Code with a Copilot Pro+ or Copilot Enterprise subscription. Here’s what you need to know and how to get started today.
Combobox vs. Multiselect vs. Listbox: How To Choose The Right One — Smashing Magazine
Combobox vs. Multi-Select vs. Listbox vs. Dual Listbox? How they are different, what purpose they serve, and how to choose the right one. Brought to you by Design Patterns For AI Interfaces, **friendly video courses on UX** and design patterns by Vitaly.
How to maximize GitHub Copilot’s agentic capabilities
A senior engineer’s guide to architecting and extending Copilot’s real-world applications.
Google’s AI advantage: why crawler separation is the only path to a fair Internet
Google's dual-purpose crawler creates an unfair AI advantage. To protect publishers and foster competition, the UK’s Competition and Markets Authority must mandate crawler separation for search and AI.
Practical Use Of AI Coding Tools For The Responsible Developer — Smashing Magazine
AI coding tools like agents can be valuable allies in everyday development work. They help handle time-consuming grunt work, guide you through large legacy codebases, and offer low-risk ways to implement features in previously unfamiliar programming languages. Here are practical, easy-to-apply techniques to help you use these tools to improve your workflow.
Run untrusted code with Vercel Sandbox, now generally available
AI agents need secure, isolated environments that spin up instantly. Vercel Sandbox is now generally available with filesystem snapshots, container support, and production reliability.
Introducing Moltworker: a self-hosted personal AI agent, minus the minis
Moltworker is a middleware Worker and adapted scripts that allows running Moltbot (formerly Clawdbot) on Cloudflare's Sandbox SDK and our Developer Platform APIs. So you can self-host an AI personal assistant — without any new hardware.
From pixels to characters: The engineering behind GitHub Copilot CLI’s animated ASCII banner
Learn how GitHub built an accessible, multi-terminal-safe ASCII animation for the Copilot CLI using custom tooling, ANSI color roles, and advanced terminal engineering.
Think outside of the box—with Claude and FigJam
With the Figma MCP app in Claude, powered by Anthropic, designers, developers, and product managers can now create AI-generated FigJam diagrams.
Power agentic workflows in your terminal with GitHub Copilot CLI
Explore the GitHub Copilot CLI and try interacting with Copilot directly from your terminal.
Build an agent into any app with the GitHub Copilot SDK
Now in technical preview, the GitHub Copilot SDK can plan, invoke tools, edit files, and run commands as a programmable layer you can use in any application.
Beyond Generative: The Rise Of Agentic AI And User-Centric Design — Smashing Magazine
Developing effective agentic AI requires a new research playbook. When systems plan, decide, and act on our behalf, UX moves beyond usability testing into the realm of trust, consent, and accountability. Victor Yocco outlines the research methods needed to design agentic AI systems responsibly.
A cheat sheet to slash commands in GitHub Copilot CLI
Run tests, fix code, and get support—right in your workflow. Stay focused and let Copilot handle the busywork.
AI-supported vulnerability triage with the GitHub Security Lab Taskflow Agent
Learn how we are using the newly released GitHub Security Lab Taskflow Agent to triage categories of vulnerabilities in GitHub Actions and JavaScript projects.
Human Native is joining Cloudflare
Cloudflare acquires Human Native, an AI data marketplace specialising in transforming content into searchable and useful data, to accelerate work building new economic models for the Internet.
Building an agentic memory system for GitHub Copilot
Copilot’s cross-agent memory system lets agents learn and improve across your development workflow, starting with coding agent, CLI, and code review.
Community-powered security with AI: an open source framework for security research
Announcing GitHub Security Lab Taskflow Agent, an open source and collaborative framework for security research with AI.
Adapting the Facebook Reels RecSys AI Model Based on User Feedback
We’ve improved personalized video recommendations on Facebook Reels by moving beyond metrics such as likes and watch time and directly leveraging user feedback. Our new User True Interest Survey (UTIS) model, now helps surface more niche, high-quality content and boosts engagement, retention, and satisfaction. We’re doubling down on personalization, tackling challenges like sparse user data […] Re
Introducing: React Best Practices
We've encapsulated 10+ years of React and Next.js optimization knowledge into react-best-practices, a structured repository optimized for AI agents and LLMs.
What AI is actually good for, according to developers
AI is designed to help you do what you love most, not replace your expertise. Discover how developer feedback and real-world experience are shaping AI coding tools that keep you in control.
Want better AI outputs? Try context engineering.
Learn how custom instructions, reusable prompts, and custom agents help GitHub Copilot deliver more accurate results.
Building the Universal Commerce Protocol (2026) - Shopify
Under the hood of UCP: the open protocol powering agentic commerce, enabling AI agents to discover, negotiate, and transact with any merchant.
Why AI is pushing developers toward typed languages
AI is settling the “typed vs. untyped” debate by turning type systems into the safety net for code you didn’t write yourself.
Penpot Is Experimenting With MCP Servers For AI-Powered Design Workflows — Smashing Magazine
[Penpot](https://penpot.app/?utm_source=SmashingMagazine&utm_medium=Article&utm_campaign=MCPserver) is experimenting with MCP (Model Context Protocol) servers, which could lead to designers and developers being able to perform tasks in Penpot using AI that’s able to understand and interact with Penpot design files. Daniel Schwarz explains how [Penpot MCP](https://github.com/penpot/penpot-mcp) serv
How we made v0 an effective coding agent
v0’s composite AI pipeline boosts reliability by fixing errors in real time. Learn how dynamic system prompts, LLM Suspense, and autofixers work together to deliver stable, working web app generations at scale.
Cooking with constraints: A designer’s framework for better AI prompts
Design and cooking share a truth: Preparation determines the outcome. Structured prompts turn AI from guesswork into a reliable design partner.
Rewriting pycparser with the help of an LLM - Eli Bendersky's website
Thoughts on starting new projects with LLM agents - Eli Bendersky's website
Agentic AI, MCP, and spec-driven development: Top blog posts of 2025
Explore the GitHub Blog’s top posts covering the biggest software development topics of the year.
WRAP up your backlog with GitHub Copilot coding agent
An easy-to-remember acronym, WRAP will help you write effective issues, refine your instructions, and get the most out of Copilot coding agent.
Giving Users A Voice Through Virtual Personas — Smashing Magazine
Turn scattered user research into AI-powered personas that give anyone consolidated multi-perspective feedback from a single question.
AI SDK 6
Introducing agents, tool execution approval, DevTools, full MCP support, reranking, image editing, and more.
Inside the feature store powering real-time AI in Dropbox Dash
The feature store is a critical part of how we rank and retrieve the right context across your work.
ChatGPT-s rivals, Kwai-s quiet rise- the top Internet services of 2025
AI competition intensified in 2025 as ChatGPT gained strong challengers. Instagram climbed, X declined, and platforms like Shopee, Temu, and Kwai reshaped global Internet usage. Our 2025 DNS data shows how Internet patterns evolved.
Hard problems are still hard: A story about the tools that change and the work that doesn't
Figma designer–turned–product manager Nikolas Klein worked on building prototyping tools for seven years. Then AI changed the game.
The future of AI-powered software optimization (and how it can help your team)
We envision the future of AI-enabled tooling to look like near-effortless engineering for sustainability. We call it Continuous Efficiency.
Introducing the Agentic Commerce Suite: A complete solution for selling on AI agents
Our Agentic Commerce Suite gets your business agent-ready. It enables you to sell on AI agents by making your products discoverable, simplifying your checkout, and allowing you to accept agentic payments via a single integration.
Updates to AI credits in Figma
We’re introducing new ways to track your AI credit usage in Figma, and soon you’ll have more ways to buy additional AI credits.
Speed is nothing without control: How to keep quality high in the AI era
AI can help you build faster than ever, but it can also produce bugs, issues, and problems. Use these strategies to keep your speed without losing control of your code.
The new identity of a developer: What changes and what doesn’t in the AI era
Discover how advanced AI users are redefining software development—shifting from code producers to strategic orchestrators—through delegation, verification, and a new era of AI-fluent engineering.
How UX Professionals Can Lead AI Strategy — Smashing Magazine
Lead your organization’s AI strategy before someone else defines it for you. A practical framework for UX professionals to shape AI implementation.
Tangle: An open-source ML experimentation platform built for scale (2025) - Shopify
Tangle saves months of compute time, makes every experiment automatically reproducible, and allows teammates to share computation without coordination.
Beyond The Black Box: Practical XAI For UX Practitioners — Smashing Magazine
Explainable AI isn’t just a challenge for data scientists. It’s also a design challenge and a core pillar of trustworthy, effective AI products. Victor Yocco offers practical guidance and design patterns for building explainability into real products.
How to use GitHub Copilot Spaces to debug issues faster
Follow this step-by-step guide to learn how to debug your issues using GitHub Copilot Spaces and Copilot coding agent.
Your stack, your rules: Introducing custom agents in GitHub Copilot for observability, IaC, and security
Use partner-built Copilot agents to debug, secure, and automate engineering workflows across your terminal, editor, and github.com.
