tshmieldev retweeted
Yesterday was all about animations. Today, let's talk multi-threading and Worklets 0.13!
We improved startup time, ported Babel plugin to Oxc, added stable networking APIs and, most importantly, opened an RFC to upstream our multi-threading code to RN.
More in the thread 🧵👇
tshmieldev retweeted
🦀 Shoutout to @tshmieldev, who ported our Worklets Babel plugin to Oxc. You can use it to save a lot of bundling time. It's available both as a Babel wrapper and standalone.
Here's how much CPU time it saves when bundling our example app.
docs.swmansion.com/react-nat…
tshmieldev retweeted
I'll be speaking at ReactCon Berlin this October!
If you're crazy about low-level stuff like me and want to know just how much overhead the JavaScript layer adds in React Native apps, don't miss my talk!
tshmieldev retweeted
Best CI use case for @typesafeai's Jev:
Enforce the rules a linter can't
every repo has rules that only live in review comments and onboarding docs
softlint puts them in a json file and checks every pull request against them, flagging the exact line
any language, ~$0.001 per PR
it's open source, link below 👇
Jev vs LLMS for clasifying tweets in your feed!
Try it out: github.com/tshmieldev/sharp
Make sure to drop a star!
It's open source!
github.com/tshmieldev/sharp
Make sure to drop a star if you find it useful <3
tshmieldev retweeted
Trying to visualize the benefit react-native-worklets
It's not super readable and needs polishing but it looks amazing already 🤩
This simulation is actually executing the code snippet, we could add it for free on every Worklets' docs page explaining how a given API works 🤓
tshmieldev retweeted
Is this Native or React Native? Both.
Core Animation is coming to layout animations in React Native Reanimated.
Stay tuned!
I made an extension that removes slop from your feed
github.com/tshmieldev/sharp
tshmieldev retweeted
This "benchmark" gives additional warm up to agent-device, but gives no such thing to Argent, the following is literally copied from this PR:
if tool == "agent-device":
prepare_agent_device_daemon
github.com/software-mansion/…
agent-device delivers more finished work for every dollar.
Benchmarked on @swmansion’s AppControlBench with GPT-5.4-mini and Haiku 4.5:
→ ~4× more completed tasks/$
→ 94.2–98.3% task completion
→ 2% faster with GPT-5.4-mini; 26% slower with Haiku 4.5 than alternative











