Thoughts on software engineering and performance
I ran the /stream bench after official Fizz and an immediate HTML shell flush. The result: 1ms TTFB versus Next.js at 12ms, with first content and throughput still tied.
I built rari.build using rari itself. The result: 0.12ms server-side processing, sub-500MB memory, and a server that handled the Hacker News front page without scaling up.
I added proper app router support, SSR capabilities, and correct RSC semantics to rari. Here's what I shipped and what it means for performance.
Introducing rari: a React Server Components framework with a Rust runtime delivering 4x faster performance than Next.js.
Comparing React and Nuxt server components: how their architectural differences impact performance, bundle sizes, and deployment.
Implementing persistent search history in React Native with AsyncStorage. Real-world code from my Bluesky contribution.
Applying David Goggins' accountability mirror to software development for honest self-assessment and technical growth.