I'm working on https://www.fontofweb.com because design inspiration platforms don’t give enough real material to work with.

Most sites fall into extremes: Dribbble leans toward polished mockups that never shipped, while Awwwards and Mobbin go heavy on curation. The problem isn’t just what they pick — it’s that you only ever see a narrow slice. High curation means low volume, slow updates, and a bias toward showcase projects instead of the everyday, functional interfaces most of us actually design.

Font of Web takes a different approach. It’s closer to Pinterest, but purely for web design. Every “pin” comes with metadata: fonts, colors, and the exact domain it came from, so you can search, filter, and sort in ways you can’t elsewhere. The text search is powered by multimodal embeddings, so you can use search queries like “minimalist pricing page with illustrations at the side” and get live matches from real websites.

What you can do:

natural language search (e.g. “elegant serif blog with sage green”)

font search (single fonts, pairings, or 2+ combos, e.g https://fontofweb.com/search/pins?family_id=109 , https://fontofweb.com/search/pins?family_id=135 )

color search/sorting (done in perceptual CIELAB space not RGB)

domain search (filter by site, e.g. https://fontofweb.com/search/pins?domain=apple.com, https://fontofweb.com/search/pins?domain=blender.org )

live website analysis (via extension — snip any part of a page and see fonts/colors instantly, works offline)

one-click font downloads

palette extraction (copy hex codes straight to clipboard)

private design collections

Appreciate feedback into the ux/ui, feature set and general usefulness in your own workflow

You’ve done an incredible amount of work — I’m definitely going to try out the Zed integration.

I’m curious though, how significant do you think it is for the agent to have semantic access through Tree-sitter?

Also what model have you had the most success with ?

Kindly let me know how your experience about Zed integration, I have done the ACP integration and merge to upstream Agent Client Protocol spec from Zed. The integration experience is quite magical, honestly. It's working in Zed , though for tool calls I'm still improving https://github.com/agentclientprotocol/agent-client-protocol...

Thank you for your very kind-words. I love building and agentic coding is my current curiosity.

> I’m curious though, how significant do you think it is for the agent to have semantic access through Tree-sitter?

For this, I'm really not sure, but since the start of building VT Code. I just had this idea to use tree-sitter to assist the agent to have more (or faster/more precise) semantic understanding of the coding, instead of relying them to figure out themself. For me, naively I think this could help agent to have better language-specific and accurately decision about the workspace (context) that they are working. If not having tree-sitter, I think the agent could eventually figure out itself. For this aspect, I should be research more on this topic. In VT Code, I included 6 language: Go, Python, Rust, TypeScript, Swift... ) via rust-binding crates, mostly when you launch the vtcode agent on any workspace, It will show the main languages in the workspace right way.

> Also what model have you had the most success with ?

I'm having mainly limited-budget so I can only use OpenRouter and utilize its vast amount models support. So that I can prototype quickly, for different use-cases. For VT Code agent, I'm using mainly x-ai/grok-code-fast-1, in my experience, it most suit for building VT Code agent it self because of speeds, and versatile in function calling and have good instruction following. I also have good successes with x-ai/grok-4-fast. I have not tried claude-4.5-sonnet and gpt-5/gpt-5-codex though. I really love to run benchmarks for VT Code to see how it perform in real world coding task, I'm aiming for Aider polygot bench, terminal-bench and swe-bench-lite, it is in my plan for now in my GitHub issues.

For VT Code itself, I instruct it to strictly follow system-prompt, in which I take various inspiration from Anthropic, OpenAI and Devin guide/blogs on how to build coding agent. But, for a model-agnostic agent, the capability to support multi providers and multi models is a challenge. For this I think I need help. I'm fortunately to have support from open-source community suggesting me to use zig, I have had good success with it so far, for implement LLM calls and implement the /model picker.

Overall in my experience building VT, the most important aspect of effective coding agent is context engineering, like all big-lab has research. A good system prompt is also very important, but not context is everything. https://github.com/vinhnx/vtcode/blob/main/prompts/system.md

// Sorry, English is not my main language, so pardon the typo and grammar. Thank you!

Could someone explain why turning off gravity makes it path so uniform ?

The system is periodical; not counting the absolute angle but the velocities, angular velocities, and the angle at the joint are all periodical functions.

When the angle at the joint is 180° the Energy and Angular momentum determines v_1 and v_2, the velocities of the masses. It is known that Energy and Angular momentum both conserve.

Therefore the system will play out the same after states when the inner joint is 180°, and all the parameters will be periodical between 2 such states.

You can observe this in the demonstration in TFA: set gravity to 0, and observe how the graph rotates, especially the furthest points from the origin (where the angle at the joint is 180°).

edit: there are probably 2 different solutions for when the joint angle is 180°? I've found a graph where one furthest place from the origin is really pointy, and then the other is rather round.

I’ve been working on https://fontofweb.com, a search engine for real-world web design.

Most design inspiration sites lean heavily on curated mockups (Dribbble) or award-winning showcases (Awwwards, Mobbin). That makes them polished, but they don’t reflect what most production sites actually look like. Font of Web takes a different approach: it sources directly from live websites, and the community can clip specific elements instead of entire pages. That means you can browse navbars, pricing cards, dashboards, etc., not just full screenshots.

Each clip is enriched with metadata (fonts, color palettes, original domain). Search works across that metadata, natural language queries (“minimalist fintech dashboard”), and even visual similarity — so you can find results either by text or by image.

There’s also a Chrome extension to snip and save from any site.

I’d like to hear from designers and frontend engineers: is this useful in your workflow? Anything obviously missing?

What if we shifted our focus entirely from the source of information to how useful and accurate it is?

I can't see how the prevalent value system could avoid being "sapio-supremacist" ? is "future proof" to include intelligences that are artificial but whose "sentience" is otherwise human equivalent or "greater"

Setup web forums, include ToS which indicates service is denied using automated tools.

People have been charged with telecommunication related crimes for hscking, and the tos csn indicate access denial.

This gated access won't stop AI, but will make account usage illegal.

People have been convicted for far less. May as well use such laws to our advantage for once.

That's the best path I can see.

I don't just want useful and accurate information. I want fun, creative, funny content that I can relate to.

Also, multiple perspectives.

That’s not the problem. The way you’re modeling it as an accuracy issue is partly solved by peer reviewed journals but that’s not what people are asking for. People are asking for authenticity and not the output of automated information generators for hire.

Is this an argument for AI? First, AI slop sucks. Second, even if it stopped sucking, it would need good input data, which it will need for a top 5 dish soap recommendation until it can do my dishes for me. Third, I want more than just useful information.

Wow, what websites are you trying

Fixed: instead of debouncing, i'm using location.hash, no network requests since hash arguments are ignored in urls.

I can't believe I was making a network request per keystroke, textbook example.

Heads up that hashes cannot be read from the server.

I'll note down the font comparison feature, there might be something to it indeed.

Yup, you can currently bookmark a website by clicking the button at the right side of the search bar, and also add it to a collection you create.

Here's my bookmarks as an example: https://fontofweb.com/2 And one of my collections: https://fontofweb.com/2/serif-heavy-sites

Thank you!