8 min read
AVIF, WebP or JPEG: Picking an Image Format in 2026
AVIF cuts photo file size 30-50% below JPEG but encodes slower
The Lab · Written in the open · Tagged performance
Lab.
Working notes from the studio. The tools I build, the workflows I keep, the mistakes I paid for.
17 entries · latest 2026-10-20
Latest entry
Profiling Liquid: How I Find What Slows a Shopify Page
Theme Inspector shows per-section render time right in Chrome DevTools Server-Timing header reveals Liquid vs network in one number Nested loops over full collections are the usual slow culprit Cached snippets with the render tag cut repeat work to near...
Read the entry
TLDR
This entry in one minute
- Theme Inspector shows per-section render time right in Chrome DevTools
- Server-Timing header reveals Liquid vs network in one number
- Nested loops over full collections are the usual slow culprit
All entries · performance
9 min read
I Found the Exact Recursion Depth That Crashes Node
Plain recursion crashed at exactly 9,275 calls deep. An iterative rewrite handled 215 times that depth in 13ms.
9 min read
Binary Search vs Array.includes() on a 50,000-Row Catalog
A one-line swap took a 50,000-row lookup from 408ms to under 1ms. I measured both methods instead of…
9 min read
I Measured content-visibility: auto on 3,000 Cards
I rendered 48 to 3,000 product cards in headless Chromium with and without content-visibility. Long pages win big,…
9 min read
The Performance Budget Every RAXXO Tool Has to Meet Before It Ships
A tool that felt instant on my desk crawled on mobile in a parking lot. The performance budget…
9 min read
Edge Functions vs Serverless for Solo Projects A Practical Comparison
Edge wins for cookie gating, redirects, AB tests at sub-20ms cold starts
9 min read
Six Months of Shopify Web Vitals Data What Actually Moved INP
INP dropped from 380ms to 140ms over six months on raxxo.shop
9 min read
Loading Skeletons That Don't Lie: 5 Patterns for Honest Perceived Performance
Content-shaped skeletons cut layout shift to zero versus 0.18 CLS for spinners
10 min read
The 1M Context Window vs Prompt Caching: When to Use Which
1M context costs full price on every query, caching cuts repeated tokens to 1/10
10 min read
Why I Standardized on Phosphor Icons Across 15 Repos (And Cut 60% of Icon Bundle Size)
Five icon systems across 15 repos created visual drift and 84KB bundles in the worst Next.js app.
8 min read
Replacing Lodash with Native ES2026: groupBy, fromAsync, toReversed, and 5 More
Lodash adds about 70KB minified, ES2026 covers most utility uses with zero dependencies