Frontend Feeds

Category

Company/Startup Blogs

12 Sources

  • Level AccessCompany/Startup Blogs7 min read

    From Policy to Practice: How to Put Your Accessibility Policy to Work

    To turn your accessibility policy into practice, translate it for each team, name owners, embed guidance in workflows, and prioritize user impact. Your The post From Policy to Practice: How to Put Your Accessibility Policy to Work appeared first on Level Access .

  • Mux BlogCompany/Startup Blogs4 min read

    Mux Robots, now in your living room

    Build richer mobile and TV video apps with the Mux React Native Player, featuring Mux Robots summaries, chapters, key moments, and remote-friendly controls.

  • Engineering at MetaCompany/Startup Blogs9 min read

    Intelligent Delivery: Making Every Unit of Compute Serve Better Ads

    Ads ranking models are growing toward LLM scale faster than the hardware to run them can keep up. An extra unit of compute has more impact on some ad requests, models, and chips than on others. The challenge is deciding where to allocate it to get the best outcome for the entire system. We built [...] Read More... The post Intelligent Delivery: Making Every Unit of Compute Serve Better Ads appeared first on Engineering at Meta .

  • Mux BlogCompany/Startup Blogs5 min read

    How a designer learned to unsee Will Smith eating spaghetti

    How Mux’s Creative Director designed and built a custom Video.js v10 skin himself, using Figma to design, Claude Code to build, and the browser to review.

  • Engineering at MetaCompany/Startup Blogs8 min read

    Your Algorithm: Giving People More Ways to Personalize Facebook Recommendations

    Your algorithm is a suite of AI-powered tools that makes Facebook recommendations more transparent and steerable, with a single editable view of the interests that shape Feed and Reels. We’re sharing how Your algorithm turns people’s natural language preferences into responsive recommendations, and how LLMs are tailored to specific tasks within the system. Traditionally, the [...] Read More... The post Your Algorithm: Giving People More Ways to Personalize Facebook Recommendations appeared first on Engineering at Meta .

  • Engineering at MetaCompany/Startup Blogs17 min read

    How We Built TLX: A Hardware-Native GPU Compiler for Production ML at Scale

    Triton Low-Level Extensions (TLX), is an extension to Triton that helps engineers optimize GPU performance while preserving Triton’s productive programming model. TLX introduces explicit warp-group and cluster-level orchestration, letting developers adopt new hardware capabilities without sacrificing velocity. TLX has demonstrated performance improvements in RecSys training, RecSys inference, and large-scale LLM pretraining. Modern GPU architectures increasingly [...] Read More... The post How We Built TLX: A Hardware-Native GPU Compiler for Production ML at Scale appeared first on Engineering at Meta .

  • CodePenCompany/Startup Blogs1 min read

    448: CSS Modules is Cool Again

    Shaw and Chris talk CSS Modules, a rather old bit of technology in the CSS-in-JS category. CodePen now supports them, which took a little work because the way they process code is a smidge unusual. We use CodePen Modules (with Sass!) here at CodePen so we knew we wanted to support them, but we got […]

  • Level AccessCompany/Startup Blogs11 min read

    Accessibility MCP Servers: What They Are and How They Work

    An accessibility Model Context Protocol (MCP) server connects AI assistants to testing engines, Web Content Accessibility Guidelines (WCAG) guidance, or existing findings, so The post Accessibility MCP Servers: What They Are and How They Work appeared first on Level Access .

  • Engineering at MetaCompany/Startup Blogs6 min read

    Extending LLDB to Debug Applications That Span CPUs and GPUs

    We built a vendor-neutral accelerator plugin architecture in lldb-server for debugging applications that span CPUs and GPUs. It separates shared debugger infrastructure from hardware-specific device control. The design represents the CPU and GPU as separate debugger processes while presenting them in one debugging session. Modern AI applications rarely execute in one place. A workload can [...] Read More... The post Extending LLDB to Debug Applications That Span CPUs and GPUs appeared first on Engineering at Meta .

  • Auth0 BlogCompany/Startup Blogsread at source

    Authentication and Authorization: What's New in .NET 11

    Learn what’s new in .NET 11 regarding authentication and authorization.

  • CodePenCompany/Startup Blogs1 min read

    447: WebMCP Rules (and We Already Have It)

    Chris & Shaw are chatting this week about the newfangled web technology, WebMCP. It involves new browser APIs like document.modelContext, meaning it needs to be a web standard, and that process is chugging along, but certainly not a done deal. That said, CodePen already has it conditionally implemented for browsers that support it, so you […]

  • Mux BlogCompany/Startup Blogs11 min read

    How I built a video sharing site with Claude Code and Mux after years away from coding

    See how Eric built ScubaTube with Claude Code, using Video.js v10 for a dive data overlay and TV mode, Mux Robots for chapters, and Mux Data for engagement.

  • Level AccessCompany/Startup Blogs10 min read

    A Guide to PDF Accessibility

    PDF accessibility means people with disabilities can read, navigate, and use a PDF, including with a screen reader, keyboard, or magnification. Among other The post A Guide to PDF Accessibility appeared first on Level Access .

  • Mux BlogCompany/Startup Blogs22 min read

    Agent experience and agent-led growth: a chat with Caelean Barnes

    Gauge CEO Caelean Barnes joins Dev Talks to explain how coding agents pick developer tools, why docs sway their choices, and how to test what agents do.

  • Engineering at MetaCompany/Startup Blogs10 min read

    F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale

    Facebook Feature Framework (F3) is Meta’s framework for authoring, compiling, and serving the machine learning features underlying ads ranking models, where a single request can require more than 20,000 features. We first presented F3 at AI @ Scale 2020. F3 reduced execution computational costs by compiling all features into a single merged graph, but that [...] Read More... The post F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale appeared first on Engineering at Meta .

  • Level AccessCompany/Startup Blogs5 min read

    AI Is Building Faster. Can Accessibility Keep Up?

    To keep accessibility moving at AI speed, catch issues in pull requests, fix them while context is fresh, and pair automation with expert The post AI Is Building Faster. Can Accessibility Keep Up? appeared first on Level Access .

  • Auth0 BlogCompany/Startup Blogsread at source

    AI ambitions are blocked by identity debt

    Japanese brands’ identity infrastructure is stalling rollout of agentic systems

  • Auth0 BlogCompany/Startup Blogsread at source

    Compliant Infrastructure, Frustrating CX

    Has a focus on regulatory compliance by ANZ brands overshadowed delivery of convenient, low-friction customer experiences?

  • Auth0 BlogCompany/Startup Blogsread at source

    Consumers are Ready for AI. Are Companies?

    Brands are in a high-stakes race to introduce AI-assisted experiences that are useful, convenient, and trustworthy—identity can be an enabler or a hindrance.

  • Auth0 BlogCompany/Startup Blogsread at source

    Moving Quickly Without Breaking Things

    As AI agents evolve from answering questions to taking action, customer identity enables the trust and transparency Europe’s consumers and regulators demand.

  • Auth0 BlogCompany/Startup Blogsread at source

    Secure and Convenient is the Real Killer App

    As customers seek out experiences that are both convenient and secure, telecom, media, and tech companies must modernize their identity platforms.

  • Auth0 BlogCompany/Startup Blogsread at source

    Trust Puts Customer Identity in the Spotlight

    As customers seek out experiences that are both convenient and secure, financial services organizations must adopt modern, external identity platforms.

  • Auth0 BlogCompany/Startup Blogsread at source

    Who Will Consumers Trust to Act, Shop, and Pay on Their Behalf?

    As AI agents evolve from answering questions to taking action, retailers need to deliver seamless experiences that are both convenient and secure.

  • Level AccessCompany/Startup Blogs6 min read

    Why Accessibility Belongs in Your CI/CD Pipeline and IDE

    By building accessibility into your CI/CD pipeline and IDE, you can find and fix issues before code merges or ships. Level CI with The post Why Accessibility Belongs in Your CI/CD Pipeline and IDE appeared first on Level Access .

  • Level AccessCompany/Startup Blogs15 min read

    Can AI Make a Website WCAG Conformant? A Comprehensive Guide

    AI can’t make a website WCAG conformant alone. Properly skilled AI can find, fix, and prevent many violations, but validating conformance requires human The post Can AI Make a Website WCAG Conformant? A Comprehensive Guide appeared first on Level Access .

  • Auth0 BlogCompany/Startup Blogsread at source

    Build an AI Wellbeing App with Fireworks and Fitbit via Auth0 Token Vault.

    Build Daily Wellbeing, a Next.js app that connects Fitbit through Auth0 Token Vault, compares your daily activity and sleep with your averages, and uses Fireworks to suggest what to try today.

  • CodePenCompany/Startup Blogs2 min read

    Chris’ Corner: Doin’ Stuff in CSS

    The float property in CSS is old school. CSS developers used to have to think about it all the time, because it was one of the mechanisms we had for building actual layouts like side-by-side columns and stuff. Those days are gone, but it doesn’t mean float as a property is entirely dead. It remains […]

  • Level AccessCompany/Startup Blogs14 min read

    What Is Assistive Technology? Types and Examples

    Assistive technology is any product, device, or software that helps people with disabilities complete daily tasks. Examples include screen readers, magnifiers, and voice The post What Is Assistive Technology? Types and Examples appeared first on Level Access .

  • Auth0 BlogCompany/Startup Blogsread at source

    Beyond Infrastructure: The New Era of CIAM Innovation

    Okta has been named a Leader in Gartner® Magic Quadrant™ for Customer Identity and Access Management.

  • SpeedCurve BlogCompany/Startup Blogsread at source

    Thanks, SpeedCurve!

    For years, at every team offsite, we’ve ended the week the same way. Somebody raises a glass and we all say "Thanks, SpeedCurve!" This is the last post on the SpeedCurve blog, so I want to raise a glass one more time. Here's the news. Over the next year, SpeedCurve is graduating. Our web and mobile monitoring is joining forces with Palo Alto Networks' back-end observability to become XCOR , a complete front-to-back observability platform. We'll fully support SpeedCurve monitoring through to late October 2027, as we migrate customers to our new platform. That sounds like an ending, but it’s not. To me, it feels like graduation day. You look back at everything that got you here, you thank the people who believed in you, and then you turn around and get really excited about what's next. So that's what this post is – a valedictory speech, minus the polyester gown and uncomfortable shoes. First, the practical stuff for SpeedCurve customers Nothing is changing for you today. Your synthetic tests are still running. Your RUM data is still flowing. Your dashboards, budgets, and alerts are exactly where you left them. And we're still here – the same people who’ve been answering your questions and digging into your weird edge cases for years. (We love the weird edge cases.) Over the next twelve months, we'll move you over to XCOR. Keep an eye on your inbox for an email from us with more details on next steps. A few things I want to be really clear about: You have time. Twelve months is a long runway, and that's intentional. You won't be doing this alone. We have a migration plan. We’ll walk you through it and support you every step of the way. Your support doesn't stop. Until the very last day, SpeedCurve will be a fully supported product. Okay, now that the practical stuff is out of the way, I want to tell a little story about where SpeedCurve came from and the epic journey we’ve been on for the past thirteen years. Where we came from SpeedCurve started in 2013 with Mark Zeman. Mark isn't a performance engineer by training. He's a designer. He built his first website in Mosaic, back when the tag was brand new, and he spent years as a teacher and creative director before he ever got started with performance. I think that's the whole secret of SpeedCurve. It was built by someone who cares just as much about how a page feels as how fast it loads, and who refuses to accept that those are two different questions. Mark also refused to accept that performance was only for developers and engineers. He was a true pioneer in visualizing data in ways that made performance accessible and interesting. (I’ve lost count of how many times SpeedCurve customers have raved to me about how much they love our charts and dashboards.) Our beta launched at the O’Reilly Velocity conference , with a shoutout on the stage from Steve Souders, who went on to join Mark and become SpeedCurve's other guiding light. If you've spent more than ten minutes in this industry, you know Steve. He literally wrote the book(s) on performance. He gave this field its vocabulary and then spent years helping us turn it into tools people actually enjoy using. More importantly, Steve connected so many of us and is the reason why the performance community remains one of the most collegial, warm, tight knit groups I’ve ever been privileged to be part of. When SpeedCurve joined Embrace last year, Mark wrote this post: Performance is about people It's very Mark: a little bit cosmic, a little bit cyberpunk, and completely sincere. In it, he made a point I've been trying to make for years without ever saying it as well. SpeedCurve looks like a bunch of charts. But it was actually always a tool for having conversations – about your users, about what makes them happy or unhappy, and about what your organization owes them. He also wrote this, which I'd like to put on a t-shirt someday: "Each of us is on a speed curve." A few years ago, at one of our offsites, Mark told our team, "I trust all of you implicitly." I've had lots of great bosses, but nobody had ever said anything like that before. It’s one of the most meaningful things anyone has ever said to me and I’ll hold it in my heart forever. SpeedCurve was never my company, but I've felt a level of commitment here that I've never felt anywhere else, and that one sentence is a big part of why. Mark, thank you. For the vision, for the trust, and for building something that so many people have cared about this much. The people who made SpeedCurve special I joined SpeedCurve in 2017 as employee #1. At the time, I explained my decision like this : "If Mark and Steve invited you to work with them, what would you say? Exactly." As someone who really, really, REALLY likes being right, I'm happy to report: I was right. Other than being a wife and mom, I've never done anything for this long. (Like a lot of folks in tech, I'm a bit of a serial monogamist. Three to five years, then itchy feet.) Almost ten years later, my feet are still happily planted. When I said yes, Steve said, "Yay, this is going to be fun!" He was right, too. It's been a ridiculous amount of fun, and that’s largely because of the people. SpeedCurve has always been a small team, and that was never a limitation. That was the whole point. When you emailed support, you got someone who'd built the thing, or someone who'd spent years explaining the thing at conferences – and often both! Some of these people have moved on to other adventures. I want everyone’s names here, on the record: Joseph Wynn , who joined SpeedCurve shortly after I did. If you love reliability, security, and accessibility, then you have Joseph to thank for leading the charge on all those fronts. And he’ll be continuing to do this with XCOR – lucky us! (He’s also a wicked surfer and is the reason why I ignored my self-doubt and climbed on a board – and stayed up! – for the first and probably last time at an offsite back in 2018.) Cliff Crocker , my work BFF, whom I've known since long before either of us came to SpeedCurve. We’ve worked together at SOASTA, then for years at SpeedCurve, and we’ll keep working together at XCOR. If you’ve ever met Cliff, then you know he’s a rock. He cares about everyone – customers, team members, end users – and he brings a massive amount of skill and experience to making awesome tools and an awesome culture. Elena Kay , who wrote so much of the code underneath everything you've ever clicked on in SpeedCurve. When she wasn’t building code, she was inspiring us all with her offline adventures – ice climbing, mountaineering, and running marathons. Andy Davies , who is one of the OG web performance gang. I don't even remember when we met. He's just always been there! What Andy doesn’t know about performance isn’t worth knowing. He’ll be continuing to build the best front-end monitoring in the world with our XCOR team. Andy Biggs (AKA Biggs, because when you're in an eight-person company, two people with the same name breaks your brain) brought a massive amount of creativity to the table. He challenged us to think about performance in fresh ways and he’s directly responsible for reimagining many of your SpeedCurve dashboards. (If you love your Vitals dashboard as much as I do, give the credit to Biggs!) That’s it – our entire team. Every time a customer learns that SpeedCurve was built and supported by just eight people, it blows their mind. It’s amazing what eight humans can do when you all care and are pulling in the same direction. As I mentioned, many of us are carrying the torch forward at Palo Alto Networks’ XCOR platform. Here’s what I’m excited about… What we're bringing with us In early 2025, on my eighth work anniversary, I tried to pin down why SpeedCurve has felt like such a good home. I landed on a handful of things : a shared vision of building tools that genuinely help people. Trust. Integrity. Humanity – meaning the kind of workplace where life can knock you sideways for a while and nobody makes you worry about your job while you heal. And fun. Lots of fun. Here's what makes me excited instead of wistful: when we started getting to know our new comrades at Palo Alto Networks, we kept finding the same things. The same vision, same goals, same commitment to trust and integrity and treating people like people. Even the same sense of humour, which – trust me – is not a given. We didn't just find a new home for our technology. We found our people. What we’re building Together we're building something bigger than any of us could build alone. The web has changed a lot since 2013. Your users don't live on a single site anymore. They move between web and mobile, and the experience they have depends on everything from the front end all the way back to the infrastructure. The problems are getting harder and more interesting. Part of what’s making them harder is AI: teams are shipping more code, and they’re shipping faster, which means more ways for things to break. We’re putting AI to work on the other side of that equation, digging through the whole stack to find the root cause when something goes wrong. Yes, it’s fuelled by AI, but ultimately humans are at the centre . We’re helping humans spend less time hunting for needles in haystacks so they can spend more time building and optimizing. And – most important – we never lose our connection to our users on the other side of the screen. That last bit is everything. For years and years, we've argued on this blog that performance isn't really about milliseconds. It's about people. The person on the spotty train connection who just needs to DO THE THING before the train goes into a tunnel. The person who gives up on your checkout in frustration and never comes back. People are central to everything we build next. One last toast To everyone who ever ran a test, set a performance budget, sent us a bug report or a feature request, came to one of our talks, or argued with us on the internet about which metric matters most: thank you. You made this the best job I've ever had. To Mark, Steve, and every SpeedCurver past and present: thank you for the epic ride. To our new friends at Palo Alto Networks: I can't wait to see what we build together. In the funny way time has, the last decade feels like a long time and like no time at all. So, glasses up, one more time... Thanks, SpeedCurve!

  • CodePenCompany/Startup Blogs1 min read

    446: Our Next-Generation of Feeds is Here

    CodePen has “feeds” for exploring content, like our Trending feed for showing off what’s popular lately all across CodePen, and our Following feed showing off the work of people you’ve chosen to follow on CodePen. Both of those require some fairly fancy technology to pull off! We’ve pulled this tech in-house, so we have more […]

  • Heroku BlogCompany/Startup Blogs3 min read

    Securing Your Stack: Recent PHP and eMudhra Updates on Heroku

    Managing critical vulnerabilities often consumes significant engineering time. When two remote code execution flaws hit PHP alongside the eMudhra root certificate compromise, Heroku managed the heavy lifting for their managed runtimes. The post Securing Your Stack: Recent PHP and eMudhra Updates on Heroku appeared first on Heroku .

  • Engineering at MetaCompany/Startup Blogs6 min read

    Catching Bugs Before They Land: JiTTesting’s Human-on-the-Loop Approach at Full Scale

    Over the last year we’ve been sharing our work leveraging large language models to catch bugs on the fly — a novel approach we call Catching Just-in-Time Tests (JiTTests). Now, we’re sharing the results and positive impact we’ve seen with Catching JiTTests deployed into production. Our paper, “Just-in-Time Catching Test Generation at Meta,” fully details [...] Read More... The post Catching Bugs Before They Land: JiTTesting’s Human-on-the-Loop Approach at Full Scale appeared first on Engineering at Meta .

  • Engineering at MetaCompany/Startup Blogs12 min read

    Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design

    This post describes the next generation of Meta Andromeda, Meta’s personalized ads retrieval engine, and how co-designing the model, system, and GPU hardware unlocks a new efficiency frontier for ads retrieval. In production, the system scores every eligible ad with near ranking-level model performance at retrieval scale and surfaces the top candidates at over one [...] Read More... The post Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design appeared first on Engineering at Meta .

  • Mux BlogCompany/Startup Blogs8 min read

    Humans and agents rejoice: Mux Robots is now GA

    Mux Robots is generally available (GA), with a stable v1 API, two new AI workflows for video structure and written content, and six months of improvements.

  • Engineering at MetaCompany/Startup Blogs18 min read

    PrivacyLib: Privacy Controls at the Data Access Boundary

    This latest deep dive into Meta’s Privacy Aware Infrastructure (PAI) takes a look at PrivacyLib. A privacy requirement must hold as data moves through Meta’s diverse data and processing systems. Historically, supporting a new requirement meant implementing the same enforcement logic separately in each system. PrivacyLib replaces this with a common library that runs inside [...] Read More... The post PrivacyLib: Privacy Controls at the Data Access Boundary appeared first on Engineering at Meta .

  • Level AccessCompany/Startup Blogs7 min read

    What Is a Digital Accessibility Platform?

    A digital accessibility platform is software an organization uses to find, fix, prevent, and report on accessibility barriers across its websites, applications, and The post What Is a Digital Accessibility Platform? appeared first on Level Access .

  • Mux BlogCompany/Startup Blogs8 min read

    What the iPhone Duo means for web development

    Which web APIs can you use on the iPhone Duo? We test Device Posture and Viewport Segments in Safari, then build a folded-screen player with Video.js v10.

  • Level AccessCompany/Startup Blogs10 min read

    How to Choose an Enterprise Accessibility Platform

    The best enterprise accessibility platform covers your digital estate, builds accessibility into design and developer workflows, pairs automation with human expertise, and proves The post How to Choose an Enterprise Accessibility Platform appeared first on Level Access .

  • CodePenCompany/Startup Blogs1 min read

    Chris’ Corner: We Heart CSS

    It warms my old heart to hear Jason Santa Maria wax poetic about modern CSS: CSS has exploded! Flexbox, grid, many kinds of viewport units, variables, and on and on. So many things we used to patch in to extend CSS or turn to Javascript for are now just part of the spec. The ease […]

  • Level AccessCompany/Startup Blogs11 min read

    What Is a VPAT? A Guide to VPATs and ACRs

    A Voluntary Product Accessibility Template (VPAT®) is a standard form for reporting product accessibility. Once completed, it becomes an accessibility conformance report (ACR) The post What Is a VPAT? A Guide to VPATs and ACRs appeared first on Level Access .

  • CodePenCompany/Startup Blogs1 min read

    445: CodePen AI is Here

    Chris and Alex are talking this week about CodePen’s biggest release since the 2.0 drop. It’s live! We’ll be working on CodePen AI forever, since it’s a fast-changing technology that regularly adds new abilities. That sounds like a trite thing to say, but we lived it as we were building this. We started out playing […]

  • Level AccessCompany/Startup Blogs11 min read

    Can AI Make a Website Accessible?

    AI generally cannot make a website fully accessible by itself. However, with proper skills, guardrails, access to vetted design systems, and human input, The post Can AI Make a Website Accessible? appeared first on Level Access .

  • Level AccessCompany/Startup Blogs5 min read

    Accessibility Needs an Authority. Why Gartner Named Us a Leader

    Level Access has been named a Leader in the inaugural Gartner® Magic Quadrant™ for Digital Accessibility Platforms, published October 5, 2026. This is The post Accessibility Needs an Authority. Why Gartner Named Us a Leader appeared first on Level Access .

  • Level AccessCompany/Startup Blogs17 min read

    Comparing AI Accessibility Tools for Building and Testing Websites

    AI accessibility tools are software that uses machine learning or large language models to help teams find, prioritize, fix, prevent, or report on The post Comparing AI Accessibility Tools for Building and Testing Websites appeared first on Level Access .

  • Level AccessCompany/Startup Blogs3 min read

    Level Access Named a Leader in the Inaugural Gartner® Magic Quadrant™ for Digital Accessibility

    Company positioned highest in Ability to Execute among all vendors evaluated STAFFORD, Va. — Oct. 16, 2026 — Level Access, a global provider The post Level Access Named a Leader in the Inaugural Gartner® Magic Quadrant™ for Digital Accessibility appeared first on Level Access .

  • Heroku BlogCompany/Startup Blogs3 min read

    Front-End Web Deep Dive: Runtime Config

    Runtime configuration has been brought to the Front-End Web Cloud Native Buildpack in a way that is standard, secure, and requires zero re-builds. By revisiting Twelve-Factor principles through the lens of modern web standards, we have created a strategy that works for any framework. The post Front-End Web Deep Dive: Runtime Config appeared first on Heroku .

  • LogRocket BlogCompany/Startup Blogs6 min read

    Traditional GPT vs. looping Transformers: Key differences

    Explore how looping Transformers reuse layers and compare with traditional GPT models in reasoning tasks. The post Traditional GPT vs. looping Transformers: Key differences appeared first on LogRocket Blog .

  • Axess LabCompany/Startup Blogs1 min read

    Webinar: AI and Accessibility – What have we learned so far?

    Learn about the potentials and be aware of the pitfalls. Join our upcoming webinar to explore the rapidly evolving intersection of AI and accessibility, from its use as an assistive technology to supporting plain language. We will share our latest findings to help you navigate the real benefits, hidden risks, and common pitfalls of relying […]