Wojciech Adamek reposted
Wojciech Adamek
Wojciech Adamek
@feathered.eu
Developer & Music producer. Creator of Sample Vault https://samplevault.ai Founder of Feathered https://feathered.eu
426 followers362 following286 posts
Wojciech Adamek reposted
Wojciech Adamek reposted
Getting closer to feature parity with other sample browsers! Except this one's free to use! #musicproduction #samplevault #beatmakers #daw #ableton #flstudio #audioproduction
v1.1.0 is live! โข Ableton Link support โ sync sample playback with your DAW โข Tempo lock โ time-stretch any sample to match your project โข Key lock โ transpose samples automatically
Discovered moonrepo yesterday and tried out @biomejs.dev . This along with @tanstack.com Start is just ๐ค All powerful tools with excellent DX, been a while since web dev was this pleasant for me.
Wojciech Adamek reposted
Started a @tanstack.com Start project today. It feels so good working with a framework that isn't actively trying to make me feel stupid. I think it's gonna become my go-to stack for a while. That and @astro.build.
Here's the gist of my settings: gist.github.com/wadamek65/b... Docs: code.visualstudio.com/docs/getsta...
See it for yourself by trying out the app: sky-scout.app It's currently free to use as I'm running a pilot of the main premium feature.
Looking for qualified leads? sky-scout.app is temporarily free to use. I'm running a pilot of the main premium feature of the app to gauge the usefulness of the product. You can sign up and use it for free for the next couple of days๐ #buildinpublic
This was quite a puzzle for me in sky-scout.app/waitlist as I want to make the app as cheap and as accessible as possible. After trying out a bunch of different models, I settled mainly on 4o-mini. Gemini Flash and Minstral results were rather mediocre.
1/ Prompt caching If you're running the same prompt, OpenAI's API offers automatic context caching for generations over 1k tokens. Put the variable part of the prompt at the end of it - otherwise caching won't take effect. This can reduce costs by 50%. platform.openai.com/docs/guides...
2/ Batching Just like caching, you can reduce the costs by 50%, except you can do that by batching multiple generations at once. This API is asynchronous and can take up to 24 hours to finish running. This is a trade-off for complexity in your app. platform.openai.com/docs/api-re...
Great tip from Dustin! And for Cursor users, there is a similar feature but Cursor is automatically aware of it. It's called Cursor rules: docs.cursor.com/context/rule... Collection of rules: github.com/PatrickJS/aw... #buildinpublic bsky.app/profile/dust...
I've been working on improving sky-scout.app/ 's algorithm and the past 2 days it has been running for over 20 projects people have submitted to me from Reddit. The results have been variable, some amazing, some bad. As I was investigating the reason, I noticed a trend.
A quick architecture diagram of sky-scout.app, for those curious. Perhaps it will inspire you to build something using the bsky's firehose๐ #buildinpublic #webdev
Finally have most of the pieces in place so this week I'm focusing on gauging the usefulness of sky-scout.app/waitlist Let me help you and try to find you customers for free! Just link me your product and I'll get back to you in a day or two with results (or lack of ๐ฌ)! #buildinpublic
React 19 has just released so I needed to refresh my memory and re-read the change log. I think I'm just gonna stick to @tanstack.com for now.
How do you stay up-to-date with the tech world? How do you stay productive and not get overwhelmed? It took me a little too long to figure this out, so I've written a short post about this: wadamek.me/blog/the-key... Iโd love to hear your thoughts and strategies!