Dynamic Image Optimization with imgproxy
summary
Images drive 42% of LCP and the conference speaker page transfers 35.7 MB with individual images up to 5 MB, so imgproxy is deployed as a dynamic on‑the‑fly optimizer that converts to WEBP and AVIF, does face detection centering, and serves one canonical source image. It runs on Cloud Foundry for Lidl & Kaufland digital leaflets, handles 5.2 billion requests a month and 751 TB traffic, uses autoscaling (min 2 max 50, scale on CPU 60% / down under 5%), caches and prewarms URLs, and saved five figures each month in CDN costs.
AI-written summary · about
chapters (6)
- 0:00Introduction and speaker background
- 4:21Schwartz Group and stack it overview
- 8:37Problems with static image handling
- 13:56Introducing imgproxy solution
- 21:22Live demo of imgproxy
- 28:10Key takeaways and closing
related talks
45mA picture is worth a thousand…kilobytes.
44mImage Optimization for eCommerce websites
45mThe image problem of the web and how to solve it…
37mImage Formats, Performance and Cognitive Load
25mDelivering Fast and Beautiful Images and Video
26mFast and Beautiful Modern Image Delivery Techniques
30mFast and Beautiful: Modern Image Delivery Techniques
43mEveryone Wins When You Optimize Media