Top all-time · #4459
canva.dev
Company blog · canvatechblog.comprogrammingwebdata
canva.dev is the company blog at canvatechblog.com, writing about programming, web and data. Since 2021, 17 different Hacker News readers have linked it 56 times, most often “From Zero to 50M Uploads per Day: Scaling Media at Canva”. It is still writing: the last post was on 2026-09-17, and you can follow it by feed.
Too few sampled comments (2) to say how often HN recommends it.?
Why HN links it
HN readers link Canva's engineering blog for posts on running a very large Git repository (about half a million translation files) and on building auto-scaling GPU training infrastructure on Kubernetes.
Summarised by Claude from a sample of 2 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
The technology behind Canva. We're sharing how we solve complex engineering problems to empower everyone in the world to design.
Most-linked posts?
- From Zero to 50M Uploads per Day: Scaling Media at Canva4 people · 6 mentions · first linked 2022-11-29
- Supporting GPU-Accelerated Machine Learning with Kubernetes and Nix4 people · 4 mentions · first linked 2022-07-05
- Enabling real-time collaboration with RSocket2 people · 5 mentions · first linked 2021-05-27
- Subscription Analytics at Scale2 people · 3 mentions · first linked 2021-07-16
- Hyperparameter Optimization with Argo2 people · 3 mentions · first linked 2021-09-16
- Deep Learning for Infinite (Multi-Lingual) Keywords2 people · 2 mentions · first linked 2023-01-30
- Simple, Fast, and Scalable Reverse Image Search2 people · 2 mentions · first linked 2022-10-17
What Hacker News says?
Hey everyone, author here, the article is a bit misleading in that .xlf files aren't really generated files, they're created through our translation pipeline by real humans. I considered them generated in the sense that they're not directly worked on by engineers who have to deal with them in the repository. The content of these translation files are snapshot in time aligned with the text in our product so simply re…
Comment linking /how-to-design-in-every-language-at-once-f2dd66a2780f
on “We Put Half a Million Files in One Git Repository, Here’s What We Learned”Submitted We Put Half a Million Files in One Git Repository, Here’s What We Learned as a story · 222 points
HN discussionJust use https://modal.com/ :) At Canva I built auto-scaling GPU infra on K8s for model training[1], and it's way too much work and operational expense to be worth building yourself. I went work at Modal because building it properly once and then distributing the solution was going to be just way better and more efficient. 1. https://canvatechblog.com/supporting-gpu-accelerated-machine...
Comment linking /supporting-gpu-accelerated-machine-learning-with-kubernetes-and-nix-7c1da8e42f61
on “The Great MLOps Hoax: Is It Just Data Engineering in Disguise?”Submitted CSS: Absolutely positioning things relatively as a story · 134 points
HN discussionSubmitted Simple, Fast, and Scalable Reverse Image Search as a story · 129 points
HN discussionSubmitted How to design in every language at once as a story · 85 points
HN discussionSubmitted Hyperparameter Optimization with Argo as a story · 27 points
HN discussionSubmitted Combining Classic and Modern: A New Approach to Camera Simulation as a story · 18 points
HN discussionSubmitted Search Pipeline: Part I as a story · 16 points
HN discussionSubmitted Migrating from Solr to Elasticsearch, and their differences as a story · 10 points
HN discussionSubmitted Supporting GPU-Accelerated Machine Learning with Kubernetes and Nix as a story · 8 points
HN discussionSubmitted Enabling real-time collaboration with RSocket as a story · 8 points
HN discussionSubmitted Testing SEO scalably without breaking a sweat as a story · 4 points
HN discussionSubmitted Recommender systems: When they fail, who are you gonna call? as a story · 4 points
HN discussionSubmitted Migrating from Solr to Elasticsearch, and their differences as a story · 3 points
HN discussionSubmitted Picking Color via Eyedropper on Web App as a story · 3 points
HN discussionSubmitted Combining Classic and Modern: A New Approach to Camera Simulation as a story · 3 points
HN discussionSubmitted How To Design In Every Language At Once as a story · 3 points
HN discussionSubmitted A Journey Through Color Space with FFmpeg as a story · 2 points
HN discussionSubmitted Deep Learning for Infinite (Multi-Lingual) Keywords as a story · 2 points
HN discussionSubmitted Search Pipeline: Part II as a story · 2 points
HN discussionSubmitted Reverse Image Search Using Perceptual Hashes and DynamoDB as a story · 2 points
HN discussionSubmitted Supporting GPU-Accelerated Machine Learning with Kubernetes and Nix as a story · 2 points
HN discussionSubmitted Infrastructure Is Distribution as a story · 2 points
HN discussionSubmitted Speeding up UI tests with CodeceptJS and Playwright as a story · 2 points
HN discussionSubmitted Picking Color via Eyedropper as a story · 2 points
HN discussionSubmitted Going Deeper with Depth Maps as a story · 2 points
HN discussionSubmitted Going deeper with depth maps to Auto Focus as a story · 2 points
HN discussionSubmitted When Recommender Systems Fail, Who Are You Gonna Call? as a story · 2 points
HN discussionSubmitted Real-Time Collaboration with RSocket as a story · 2 points
HN discussion
