Image Optimization for eCommerce websites
summary
Claims a cloud image service that does real time resizing and format switching under 50 ms per conversion, detects device and network, and serves WebP/HEIC when appropriate. Shows numbers: ~60M monthly users historically, 2–3M uploads/day, and the business rule 1 second delay costs ~7% conversions. Describes static vs dynamic tradeoffs, C implementation, AWS marketplace VM and pay‑as‑you‑go pricing ($1 per 1,000 queries), and live deployments at Nordstrom and Snapfish.
AI-written summary · about
chapters (26)
- 0:00Introduction
- 0:48Jacks background
- 1:27First project at Google
- 2:10Google Compute Cloud
- 2:31Imageshack
- 3:05Image Check
- 6:02Pinterest
- 7:40quadcopter
- 8:01eCommerce
- 9:19Infographic
- 10:43Responsive Design
- 12:17Product Images
- 16:53Dynamic Images
- 21:10Image As Your Engine
- 23:48Image Azor
- 24:27Nordstrom
- 24:51Snapfish
- 25:17ImageAzor
- 26:20Team
- 26:30Demo
- 30:30Nordstrom example
- 31:52Computer vision example
- 32:55Image Optimization API
- 34:50Object Recognition
- 39:05Policy options
- 41:10AWS Marketplace
related talks
30mReimagining the image pipeline at Squarespace
42mEngaging Ecommerce with the Visual Web
37mImage Formats, Performance and Cognitive Load
06mOvercoming the Challenges of Image Delivery
31mEngaging Ecommerce with the Visual Web
36mDynamic Image Optimization with imgproxy
25mEngaging Ecommerce with the Visual Web
45mA picture is worth a thousand…kilobytes.