Bilt.me
AI Agent that builds real Native Mobile Apps
214 followers
AI Agent that builds real Native Mobile Apps
214 followers
Build and launch production-ready native mobile apps with AI. Prototype, collaborate, and deploy faster — no coding required. Build from scratch or auto deploy Web Apps into Mobile Apps. By making mobile app creation as quick and frictionless as writing a thought, we aim to unlock a wave of innovation where anyone can turn fresh ideas into working native apps the moment inspiration strikes.
This is the 3rd launch from Bilt.me. View more
Bilt.me - Figma
Launching today
You designed the whole thing in Figma. Then you had to hire someone to rebuild it by hand, pixel by pixel, and it came out "close enough." Not anymore. Drop your Figma frames into Bilt and they become a real app for iPhone. Your spacing, your fonts, your colors, exactly. You even own the code.
Free
Launch Team
Bilt.me
ChatWebby AI
The "you own the code, dropped on GitHub" part is what makes this stand out to me — most no-code-to-app tools lock you in, so giving people a real exportable codebase is a great call. Since the app is built from the actual Figma file rather than a screenshot, how does it handle iterating after the first build? If I tweak spacing or swap a font in Figma later, does re-running the plugin update the existing app cleanly, or does it regenerate from scratch?
Bilt.me
@zain_sheikh Yeah, every change you do after won't rebuild the entire app. As long as you reimport your frames - iterating should be easy aswell. Truly as simple as developing through design :)
Serializing the design tokens and semantics instead of a flat screenshot is the right move, that's where the pixel fidelity actually comes from. The place I'd expect it to wobble is layout intent: a frame built with auto-layout gives you real constraints to map onto native, but plenty of designers hand-position everything and detach components, so there's no structure left to read. Do you derive native layout from auto-layout when it's present, and what happens with frames that are just absolutely-positioned pixels?
Bilt.me
@dipankar_sarkar Me personally, I don't really design apps using "auto-layout" either, but Bilt has done me quite well. Obviously, if there are existing layouts, Bilt uses those, but absolutely positioned objects will also show up in the app itself. I suggest you go try it out :)
jared.so
Hey, congrats on the launch, how is it different than just handing the screenshot to an LLM? Does some data of the design get passed also to Bilt? Genuine question, I think this is insane if it works out exactly as the Figma design
Bilt.me
@borrellbr Hey! Instead of it being given just the screenshot, the content of the figma frame is carefully serialized for agent to understand it the best. The serialized content includes most of the metadata, design tokens, semantics.
Congrats! I am curious about how much control does the maker have over platform-specific details? For example, can someone steer iOS and Android patterns separately, or is the agent mainly focused on getting a working native app structure from a higher-level prompt?
Bilt.me
@crystalmei Hey! The agent is focused on handling both platforms simultaneously to the best of it's ability. The rest of the platform is currently focused more on iOS compared to Android, but it's in the pipeline!
woah this looks exciting! but in android studio we have the option to integrate figma automatically. does that not code the front end now with the deep gemini integration? If not then this really could be a new genre of dev tools!
Bilt.me
@nischay_kashyap the android studio + gemini combo is great, but it's really just a coding assistant for someone who's already in android studio writing kotlin. it'll generate a screen, but you're still the developer assembling the app around it.
bilt is doing something different. you give it a figma file or just describe what you want, and it builds the whole thing into a real, native mobile app you can ship to the app store or google play :)
so less "figma to front-end code," more "idea to shipped app.", would love to hear what you'd build with it.
Fair enough that they render. The thing I'd actually watch is what happens across screen sizes: absolute coordinates map 1:1 on the fixed Figma canvas, but a native app runs on a 6.1 inch phone and a tablet, so fixed positions either letterbox or drift. When there's no auto-layout to read, does Bilt infer constraints (pin this to the bottom edge, stretch that to width) or does it place things at the literal coordinates and let them sit?