ReScript enthusiast at @envio_indexer. Create tools from the future 🧬
Georgia, Batumi
Joined March 2022
- Tweets792
- Following59
- Followers287
- Likes937
Pinned Tweet
"Parse, don't validate" caught on in JS. The encoding side is still very early, with messed-up outbound data in the majority of projects - new article on that.
Feat: @colinhacks, @EffectTS_, @Rich_Harris, Typia, and others.
A retweet would help a lot 🙏
dev.to/dzakh/encode-dont-str…
Grok Build was a huge inspiration in terms of TUIs recently.
Running an indexer locally now comes with a new terminal UI.
We rebuilt it in Rust and gave it its own thread, separate from the indexer, so it stays responsive while your chains sync.
Check out the latest release.
envio.dev/changelog
Sury v11.1.0-rc.0 is out!
- Protobuf support 5x faster than any other JS library
- Standard Schema result returns multiple issues - perfect for form libraries
- 40x faster parsing with failures
- New S.compareInput and S.compareOutput
- Default value support for S.nullish
Dmitry Zakharov retweeted
ECMAScript excitement 😉
Congrats to @acutmore from @TechAtBloomberg on advancing the Composites proposal to Stage 2 at @TC39 today 🎉
Composites are objects that are compared by value instead of identity. They are an evolution of Record & Tuple proposal.
ALT const pos1 = Composite({ x: 1, y: 4 }); const pos2 = Composite({ x: 1, y: 4 }); pos1 === pos2; // true const positions = new Set(); // the standard ES Set positions.add(pos1); positions.has(pos2); // true
Dmitry Zakharov retweeted
Now that Schema Benchmarks has been officially released for 7 months, I thought it'd be useful to look at how it's improved over that time, and how it's helped the ecosystem!
schemabenchmarks.dev/blog/7-…
A new article. Shared some thoughts and design decisions I applied when shaping Sury v11.
dev.to/dzakh/i-have-designed…
Dmitry Zakharov retweeted
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Thanks. I didn't enjoy writing much before, but after @GergelyOrosz's recent tweets, I gained an interesting insight.
Nowadays, as our job has become managing AI agents, the act of writing is the artisanry we "lost" in software development.
Sury x @nuqs47ng = A match made in Heaven!
The schemas' composability, with a neat adapter architecture from nuqs. I don't know, man; this looks just beautiful to me.
There's an interactive example in the docs by the way nuqs.dev/docs/parsers/commun…
This is what I do instead of writing articles. By the way, it's 99.1% Google's test-run conformance and 4-10x faster than Protobuf-es
From time to time, I look at a friendly competition between Valibot vs Zod. Usually, Valibot wins in many directions, but today I found its weakness.
Who could have guessed that Zod is the FASTEST library in terms of TS type performance 🔥
ALT Zod 8.5x faster than Valibot
Don't mind Sury being slower; the DX is worth it.
For @arktypeio funs, it's at 27_049 instantions which is 27x more than Zod. But very impressive, considering what the guys do.
Inspired by @ssalbdivad's Attest, I decided to add TS inference comparison to Schema Benchmarks.
I really like what @eskimojo built, and I think having an enhanced benchmark like this is underrated for the JavaScript community.
Declaring and paying monthly taxes - the most annoying thing I always had, but not annoying enough to create a Playwright script.
Automated with @bot now. I'm happy.
At this point, I'm happy that I have a secondary bank account I don't afraid to have drained 😅