@ShopifyEng
Account based inCanada
About this account
- Account based in
- Canada
- Connected via
- Web
Account-level information from X, not a live location or the device used for a specific post.
Making commerce better for everyone. Follow us for technical discussions and updates on how engineers build @Shopify. Explore open roles: https://nitter.cf/t.co/NFSvvCJBXt
Working Remotely
Joined May 2016
- Tweets5.1K
- Following707
- Followers59.8K
- Likes4.8K
#railsworld's first livestream is a can't miss-- join us as we MC the livestream at 9:30am!
Not making it to Austin? No worries, thanks to @Shopify you can still catch the Rails World Opening Keynote livestream tomorrow at 9:30 AM CT. Set a reminder and tune in here: youtube.com/live/V9SxpJpHuus
🧑💻 AI is dramatically increasing the amount of code teams can produce. But code generation is only one part of software delivery. What have teams had to change - in planning, architecture, review, testing, shipping, and collaboration - to keep pace?
At today's closed door Rails at Scale Summit, Shopify hosted the AI panel: When code is no longer the bottleneck, to hear perspectives from @rails orgs at various scales. Thanks to our participants: @Zendesk, @salesforce's fin, and Perry St.
Congratulations @MParakhin!
Thank you @Time for the recognition time.com/collection/executiv… and congratulations to the full list of honorees!
Did we mention we like building internal tools?
When we announced our move back to native, we got a lot of questions about how Helix works. Today we’re sharing more details.
• Instead of expecting a perfect first attempt, Helix breaks work into small checkpoints that must pass strict quality gates before moving forward.
• Different AI models are good at different things. Gemini has been especially effective at catching subtle visual differences during UI reviews.
• Engineer feedback goes into memory, so Helix gets more autonomous with each checkpoint. It can run overnight without approvals, and the gates stay just as strict.
Link to blog post👇
📊 Bug bounty surveillance log | August '26
• 599 vulnerability reports intercepted
• 257 new hackers onboarded
• 46 reports awarded bounties
• $186K+ distributed to hunters
Thank you, @rails — see you in Austin next week! We run the world's largest Rails monolith, and we're bringing Rails Core office hours to the Shopify Lounge. Stop by, talk scale, and explore what life at Shopify is all about. 💚
Rails World 2026 owes an enormous debt of gratitude to Event Partner @Shopify, this year. Their support is making all of the following possible:
- Hosting the very first Rails World livestream of the Opening Keynote for those who can't make it to Austin
- Helping @yukihiro_matz get to Rails World for the Day 2 fireside chat
- Co-sponsoring the Closing Party (always a highlight!)
- Co-sponsoring the Rails at Scale Summit
- Hosting the Shopify Lounge in the Sponsor area
- Covering post-production so that every session is edited and on YouTube within two weeks after the event
- They also sponsored the props this year. Because yes, this year the venue called for props. (You’ll have to see what that means onsite next week.)
If you're joining us in Austin, stop by Shopify's lounge to connect with Rails experts, unblock tough technical challenges, discuss building and growing applications at massive scale, and explore opportunities with @ShopifyEng. (Including this open role: jobs.rubyonrails.org/jobs/40…)
Thank you, Shopify!
ALT Shopify logo with text: Event Partner Announcement
Shopify is moving its mobile apps from React Native back to native. Here’s why:
shopify.engineering/back-to-…
We’ve already migrated the Shop app back to Swift and Kotlin. The team went from proof-of-concept to published native apps in just 12 weeks:
shopify.engineering/shop-app…
Step inside Shopify for the day and check out what's under the hood of Remix. 🏎️ 🏎️ 🏎️
Start your engines. The Remix Jam 2026 schedule just dropped. 🏁
Join us October 2 in Toronto for a full day of talks from the Remix team and expert builders from across the industry. We’ll dig into Remix 3, our approach to unbundling, Agentic Experience, Frames, SPAs, testing, as well as extending and building with Remix.
Explore the full schedule and get your tickets: remix.run/jam/2026
The best security work is often invisible: the incident prevented, the poisoned package caught, the deploy that keeps moving.
Shopify's 4-year, $1M partnership with Ruby Central on Ruby Shield showed what sustained investment in shared infrastructure can unlock. Now we’re carrying that work forward with the Ruby Alliance.
RubyGems.org now handles tens of billions of gem downloads a year with modern auth, verifiable publishing, and 24/7 coverage.
Thanks to Ruby Shield, our 4-year security partnership with Shopify. Here's our reflection as it wraps up: buff.ly/YwpC9qK
We use River, our AI Slack agent, to run our vulnerability remediation pipeline. It carries the work through live-state checks, CI, human decisions, merge, and verification. This process cut one of our backlogs by ~70% in 11 days. We’re sharing what worked so other defenders can build on it: shopify.engineering/river-vu…
👀 Try it!
Today we published the first release candidate for Remix 3!
- 350+ commits since beta
- Full-stack HMR
- Database tooling
- Safer routing
- Improved assets serving
- More components
- SPA support
- One remix package
Try it: npx remix@next new my-app
remix.run/blog/remix-3-relea…
Shopify Engineering retweeted
The recording is out for the ICML 2026 talk I gave with @cmazzaanthony on our continual learning work at @Shopify : Model Optimization Flywheel: Continuously Self-Improving LLMs in Production. Watch it here: icml.cc/virtual/2026/75732
Gisting is how we reap the behavioral advantages of a long prompt at the cost of a short one.
We outline our implementation on Eng Blog: shopify.engineering/gisting
🇰🇷 ICYMI: our @icmlconf talks are live now — no badge, no jet lag required. 🧵
First up: the Model Optimization Flywheel. The premise: a continual learning pipeline that turns Shopify's product expertise into robust evals, mines low-scoring conversations, critiques them, repairs them, and feeds them back into the model. Consider it a deep dive on last week's blog post.
Next: our workshop on the generative turn in search and rec with @Meta_Engineers, @TolokaAI, @Kuaishous, and Shopify's Foundation Models team. The multi-stage discriminative pipelines behind search, feeds, and e-commerce are giving way to unified generative foundation models — no more cascaded errors, better hardware utilization, optimization objectives that go well past "did they click."
The open questions we got into: are search and rec converging into one foundation model, what are the real scaling laws for discovery, and how do you stay under 50ms latency at trillion-item scale while handling cold starts, non-stationary catalogs, and fairness trade-offs.
We rebuilt our mobile end-to-end testing framework with a strict API and computer vision, and we raised test stability by a lot. The latest on our Eng Blog: shopify.engineering/mobile-e…
