Daniel Wasmus
@danielwsms
Followers
72
Following
7K
Media
2
Statuses
33
Introducing shadcn/create – Build your own shadcn/ui Customize Everything. Pick your component library, icons, base color, theme, fonts and build something that doesn’t look like everything else. Now available for Next.js, Vite, TanStack Start and v0.
465
1K
9K
Introducing Base UI v1 ✔︎ 35 unstyled UI components ✔︎ New npm package (base-ui/react) ✔︎ New website ✨ ✔︎ Configurable, composable, customizable ✔︎ Accessible, based on ARIA + WCAG https://t.co/HHGtrMZZLy
base-ui.com
Unstyled UI components for building accessible web apps and design systems.
100
199
2K
> be me, go to SF > never did a hackathon before > find the context engineering one from @Theoryvc > claude code yolo with Mixedbread > won > have a new PC
3
6
41
One question I get endlessly is: "I want to customize an LLM. Do I need to train my own model?" No. Repeat after me: The system is JUST AS IMPORTANT as the model.
9
12
109
I've heard people say "grep is all you need" for coding agents, so this research was 🤯 for me. Every single model was better with grep + semantic search!
Semantic search improves our agent's accuracy across all frontier models, especially in large codebases where grep alone falls short. Learn more about our results and how we trained an embedding model for retrieving code.
32
27
603
Semantic search improves our agent's accuracy across all frontier models, especially in large codebases where grep alone falls short. Learn more about our results and how we trained an embedding model for retrieving code.
72
108
1K
Available on the Vercel Marketplace: https://t.co/aMjikbCXrq
vercel.com
The Search API for your data. Add mulitmodal & multilingual search to your app
1
0
3
Mixedbread makes image search feel like magic. No keyword labeling, no manual annotation. Describe what you are looking for and instantly find it.
1
5
17
> be me > work on stuff > email from @vercel alerts > :attention mode: > check traffic source > art demo got featured in a newsletter > :calm mode: Whoever cooked the alerts. Amazing work.
0
0
4
There is no reason to not use this. It’s easy free lunch🥪
Better Projection Variants Improve ColBERT We introduce a rare free lunch: a tiny architectural modification that improves ColBERT performance across the board, without any real tradeoffs. The modification? Better final projections than the currently used linear projection.
0
4
25
One More (Small) Thing: Introducing mxbai-colbert-edge-v0 17M and 32M. They are are the result of an easily reproducible way to train ColBERT models from scratch. They're strong, too: the 17M variant would rank first on the LongEmbed leaderboard for models under 1B parameters.
5
23
128
Better Projection Variants Improve ColBERT We introduce a rare free lunch: a tiny architectural modification that improves ColBERT performance across the board, without any real tradeoffs. The modification? Better final projections than the currently used linear projection.
5
16
93
> use mixedbread > upload thousands of files via cli > start searching in minutes > look inside > its multimodal > its multilingual > :happy:
No dude all you have to do is put all 25M of your company PDFs and files through a document API then put them in a Vector DB then use RAG to find things and feed them to a custom fine tuned LLM sandbox and you’ll be able to ask it questions. It’s AI bro.
0
2
12
You can now add AI Search to your @astrodotbuild Starlight 🌟 documentation "With Mixedbread, search results instantly felt more relevant, contextually accurate, and tailored to the query that the user entered." - @imax153(@EffectTS_) Read our guide how you can get started 👇
2
6
15
