tgindex
Wix Engineering

Wix Engineering

Статистика

Wix Engineering official channel

Последний пост
22 июл.
Последнее чтение
15 авг.
Постов за неделю
0
Всего постов
20
Тип
открытый
Язык
und
Категория
Новости и СМИ (по похожим)
В каталоге с
14 авг.
Подписчики
373
0 за 1 дн.
Сутки
0
0,00%
Неделя
 
Месяц
 
Просмотров на пост
89
20 постов
Вовлечённость
23,9%
к подписчикам
Постов в день
0,0
всего 20
Упоминаний
0
каналов
Охват размещения
оценка
1/24сутки в ленте
71
1/48двое суток
81
1/72трое суток
87

Оценка по просмотрам недавних постов: пост набирает почти всё за первые сутки.

Посты

  • What happens when AI starts generating, evolving, and deploying code on behalf of users - at Wix scale? The existing developer platform was no longer enough. In this talk, Gilad Bison and Dor Chaouat will show how one ambitious AI project pushed Wix to rethink the entire lifecycle of cloud-based development. Through the architecture of Dev Machine, Wix Hosting, Code Store, IDE as a Platform, and Wix CLI, you’ll learn how to design a platform for AI-generated code, securely isolate and execute untrusted workloads, connect every stage from development to deployment, and turn infrastructure built for one product into reusable capabilities for an entire organization: https://youtu.be/43H6NGQEXis?si=Ed9HrraN3O8yLvnW @WixEng ⚡️

  • In our latest Wix Engineering newsletter, you’ll learn how AI agents can catch and fix their own mistakes, how TDD can guide AI-written software, why larger context windows may reduce accuracy and increase costs, and how agents reason across code, logs, databases, and documentation. You’ll also explore the hidden economics of AI coding tools and the engineering skills that matter most when AI writes the code. Read here @WixEng ⚡️

  • Production issues. Technical debt. Migrations. New features. When everything competes for 👀 attention, developers lose focus, make more mistakes, and burn out. In this great talk, Dalia Simons presents a simple weekly rotation between Guardians, Renovators, and Creators - that reduces context switching and protects deep work. Watch to learn how clearer ownership can help teams stay focused, move faster, and ship better: https://youtu.be/IoCXn9nEr-M?si=x3qRuiisMJNdTSRZ @WixEng ⚡️

  • If AI writes the code, what’s left for engineers? 🤔 That shift changes the job. The real advantage is no longer just writing code, but asking better questions, understanding tradeoffs, owning the outcome, and developing the taste to know what should be built. Tom Enden’s article explores the engineering skills that matter when code is no longer the hardest part. Don't miss this one: https://www.wix.engineering/post/the-end-of-determinism-what-s-left-for-engineers-when-ai-writes-the-code @WixEng ⚡️

  • The more rules you add to an AI agent, the more fragile 🚩 it can become. Prompts get longer. Context gets heavier. And the agent still ignores constraints, repeats mistakes, or produces output that looks impressive - until it hits production. Yoav Abrahami proposes a different architecture: Feedback-Driven Agents. Instead of expecting perfection on the first try, deterministic validators catch structural errors, inject the exact guidance needed, and send the agent back to repair its own work. The promise: fewer broken outputs, lower costs, faster iteration - and a much shorter path from an AI demo to a reliable production system. Read here @WixEng ⚡️

  • Real incidents. Real lessons. A stronger 💪 production culture. In this great session "How to Win in Production" by Guy Nahum, we break down the failure patterns behind real incidents - and the practices that helped us at Wix reduce noise, improve on-call, and learn without blame. You'll learn practical rules for better production - and a clearer view of where AI can responsibly help. Watch: https://youtu.be/CaNuUuMsh1k?si=3JFXuMoHFLFyNaPB @WixEng ⚡️

  • What can a classic TDD book 📙 teach us about building software with AI? Growing Object-Oriented Software, Guided by Tests by Steve Freeman and Nat Pryce was written long before coding agents became part of everyday development - but its core ideas may be more relevant than ever. The real challenge with AI-written software is not generating code. It’s knowing whether the code reflects the right intent, handles the right edge cases, and can evolve without weakening the system around it. In this great session by Mantas Indrašius, you'll explore how TDD can become a practical feedback mechanism for AI-assisted development: helping coding agents verify their changes, expose wrong assumptions, iterate toward correctness, and produce more maintainable systems. Watch: https://youtu.be/2cWNrU40CAs?si=ickGpZAB2XnH4qcW @WixEng ⚡️

  • Channel photo updated

  • Your AI writes code ⚡ faster than ever. So why is shipping still the 🚩 hardest part? Because AI doesn't fail at writing code. It fails at understanding context and proving it's correct. At our next meetup (TLV) we'll tackle the two biggest engineering challenges of AI-first development: 🧠 How do you give AI the context your team has - but your repository doesn't? ⚙️ How do you redesign CI - when AI agents generate more code than humans ever could? Expect practical patterns and ideas you can apply to your own engineering team the next day! Join us: https://www.wix.engineering/beyond-the-prompt-engineering @WixEng ⚡️

  • How do you build a remote browser platform that scales to 🚀 massive parallel execution - without becoming the bottleneck itself? That's exactly the challenge David Gahnassia and his team tackled while building the remote browser architecture behind Sled3, our next-generation E2E testing framework. In this great talk, you'll get an inside look at the production architecture - from Kubernetes and the proxy layer to scaling strategies, design trade-offs, and the lessons we learned building a platform that can run E2E tests at scale. This isn't a talk about writing better tests. It's about building the infrastructure that makes large-scale testing possible. Watch: https://youtu.be/3KnYcy_1LFM?si=_0nL0OIWG74GwLEh @WixEng ⚡️

  • Why do AI agents 🚩 fail in production? Because the answer rarely lives in a single repository. In his new article, Guy Bary shares how we built Wix Octocode - an AI research engine that connects code, logs, databases, documentation, and organizational knowledge to uncover the real root cause before taking action. You'll learn why search isn't enough for modern AI agents, how multi-agent architectures improve research, and why context engineering is becoming one of the most important challenges in AI-native software development. Read here: https://www.wix.engineering/post/how-we-built-the-brain-behind-our-self-healing-system-context-retrieval-at-org-scale @WixEng ⚡️

  • AI coding breaks 🚩 down not because of bad prompts, but because it forgets your project's decisions. Yoav Abrahami's Design-Log Methodology fixes this by giving AI a permanent, versioned history of architectural choices - keeping it aligned, reducing drift, and turning it into a true design partner instead of just a code generator. Watch his great session "Stop Prompting. Start Logging: Breaking the Context Wall in AI-Driven Development" https://youtu.be/fLOlkjRl6lo?si=RbPGJiI7sz3oD-QI @WixEng ⚡️

  • Proprietary AI models, zero-downtime database migrations and 13 years of engineering community. The latest Wix Engineering newsletter covers our new proprietary AI models powering Wix Harmony and Base44, how we migrate production databases with zero downtime, what 250 AI agent evaluations taught us, why documentation is becoming AI infrastructure - and more! Read the full edition: https://www.wix.engineering/so/62PyIEe7N @WixEng ⚡️

  • 13 years ago, we booked a room, ordered 🍕 pizza, and invited developers to talk about real engineering problems. That simple idea became "Meetups at Wix". The goal was never recruiting or marketing. It was always about engineering. We wanted to share what we were learning while building and scaling one of the largest web platforms in the world. Working on infrastructure that serves hundreds of millions of websites exposes you to challenges you simply won't find in textbooks - and those lessons were too valuable to keep to ourselves. Thirteen years later: 🎤 250 meetups 👥 15,000 developers attended 🍕 5,500 pizza trays served 💡 620 engineering talks ⭐ 975 reviews with an average rating of 4.5/5 Technology changes. Languages change. AI changes how we build software. But one thing hasn't changed in 13 years: the fastest way for our industry to move forward is still engineers learning from other engineers. @WixEng ⚡️

  • The biggest shift in software engineering isn't a new programming language. It's a new way of 🧠 thinking. In this great talk, Tom Enden explains why AI is pushing us beyond deterministic programming into a probabilistic world, what that means for modern engineers, and how to build reliable systems when your teammate is an LLM. Watch: https://youtu.be/VIYnXUQkKsU?si=3CGfvEq6XkJRG7_v @WixEng ⚡️

  • AI is changing how we build software and solve problems. But some of the most important 🚩 aspects of engineering - still depend on us. In this talk, Aviva Starovolsky-Peisach explores some of the most common debates around diversity and bias in tech - not with slogans, but with personal stories, real data from Wix, and practical techniques you can apply immediately. Whether you're a team lead, interviewer, or individual contributor, you'll walk away with practical ideas for building stronger teams and making better decisions - because the best engineering organizations are defined not just by their technology, but by the culture and people that drive it: https://youtu.be/u_aOrsZYzq4?si=Q4DiVg7tgzI6IH_o

  • 🚶‍♂️‍Building AI agents is easy. 🏃‍♂️‍Proving they work is hard. In this great session, Or Goldreich presents EvalForge, a platform for systematically evaluating AI agents and turning subjective impressions into measurable engineering metrics. ✅ Learn how to test agent behavior, measure reliability, detect failure patterns, and understand why AI evaluation is becoming as important as software testing itself. Because in production, "it looks good" is not a metric. Watch (Hebrew): https://youtu.be/eAhmj7weHss?si=HSjJEzr6PQzedDwL

  • 8,000 Subscribers 🚀 - By Developers, for Developers! This week, Wix Engineering's YouTube channel crossed 8K subscribers. The channel was created with a simple goal: sharing engineering knowledge with the broader tech community. Most of the content comes directly from the meetups, conferences, and community events we organize, where Wix developers step onto the stage to share real-world lessons, architectures, failures, breakthroughs, and the technologies they build every day. Since launching the channel, we've reached: 📺 732.3K views ⏰ 91.2K watch hours 👀 2M impressions Onward 💪 to the next milestone! https://www.youtube.com/@Wix-Engineering @WixEng ⚡️

  • Before a ticket reaches engineering, an AI system has already investigated it, generated a fix, and opened a reviewed PR. Sounds futuristic? It's already running in production. In this talk, Israel Zablianov explains the architecture, context engineering, and safeguards behind autonomous bug resolution systems. If you're interested in AI agents, developer productivity, and the future of software engineering, this one is worth your time (Hebrew) https://youtu.be/3BNoppi6qcs?si=IZ_78GC0hDp0A6H0 @WixEng ⚡️

  • What happens when your platform owns the complexity instead of your engineers? (🗓 June 25) At CoffeeJug, Borys Bulyha will share how we, at Wix, built a platform that powers ~3,000 services, enables hundreds of daily production deployments, and made adopting Java feel almost like flipping a switch. A fascinating look at platform engineering, developer productivity, AI-native development, and the engineering decisions behind one of the industry's most interesting technology shifts. Join here: forms.gle/ySR6rq7hLhiDoAb57 @WixEng ⚡️