Twitxy Saved

Shanon Jackson

Verified account This account is verified. Learn more

@Shanshrew

Principal Engineer @Atlassian / TSConf Speaker / ViteConf Speaker / Chief Yapper
New Zealand Joined March 2019
27 Following 1.1K Followers 144 Posts

Latest posts 5

Shanon Jackson Verified account @Shanshrew
Fairly loaded. SSR decouples your application size from user-experience metrics related to paint.

If your pages are SSRed as your bundle size grows over time you'll notice that your user-experience metrics related to paint stay almost entirely stable.

When smaller apps use SSR they notice less of a performance difference because their applications are smaller.
Ronan Berder (SF + SG) Verified account @hunvreus 2d
New rule: never use React SSR.

I convinced myself that things would be different this time, but SSR still sucks. It adds a ton of complexity, increases your bill, and solves no problems (for apps).
View quoted post on X
Shanon Jackson Verified account @Shanshrew
Honoured. End of 2000+ hours of work.
VoidZero Verified account @voidzerodev Oct 14
Your linter, bundler, and dev server all parse the same JS over and over. No matter the parsing cost, they all pay it.

@Shanshrew made it 2x faster on 95% of files, landing in Oxc soon.
Meaning @vite_js, Rolldown and Oxlint all benefit.

Learn more during his talk at @ViteConf
Thu 15 Oct, 2pm

viteconf.org
View quoted post on X
Shanon Jackson Verified account @Shanshrew
Speaking at ViteConf 2026 on my Research. Hopefully will do a better job than any blog could have ever done.

Hope to see you all on Oct 15th.

There's two versions I want to do of this talk; One targeted towards web developers (This one), And another targeted towards systems programmers which cover ILP, OoO pipelining, Advanced SIMD + more I'd love to do next year.
0 replies 0 reposts 10 likes 389 views View on X
Shanon Jackson Verified account @Shanshrew
1 Year of Research coming to an end. 8 Months seeing no results, Almost quitting twice. November/December of last year finally seeing a very narrow path to a universal 2x perf improvement on every modern JS parser which would take me another 4 months to build.

This will land in oxc_parser fairly soon speeding up OXLint, OXFmt, Vite, Rolldown, Deno (oxc consumer) almost immediately.

In the long term it's likely that it will land in V8/SpiderMonkey/Bun/JavascriptCore/Typescript/+ Every other JS parser implementation

The wins come from a few key elements, obviously 1 novel "SIMD For Context Sensitive Languages" the other key elements take it from a 20% win e2e to a 80-200% win (99% of files will see a 80-100% perf improvement).

These key elements translate to any AST design and thus will speed up not only all JS tooling but every major browser when complete.