I built a 1.8MB native app with self-built UI, vision and AI libraries
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Want to reach these readers? Sponsor an issue.
HN · launched Dec 1, 2025 · 15 upvotes (top 40%) · 0 views
Niche: desktop ai assistants · 275 launches, 15 in the last 7 days ▼ -25% · Explore this niche →
Category and tags
- Horizontal
- Dev tools
- Developer
- AI feature
- Hobby / open-source project
Visit website ↗ Source listing ↗
More like this: all 275 launches in desktop ai assistants →
Could you build this?
No Hand-rolling a custom UI engine, computer vision library, and AI inference runtime into a 1.8MB native binary requires extreme low-level systems engineering, math, and binary size optimization.
What it would actually take: This architecture requires C/C++ or Zig with custom software rendering (or direct bare-metal GPU shaders via Metal/Vulkan), handwritten matrix math routines, and custom quantized neural net kernels without third-party frameworks like PyTorch or OpenCV. The hardest part is stripping standard runtime bloat while implementing custom convolution, rasterization, and tensor graph execution. It requires rare, elite expertise in embedded systems, graphics programming, and compiler-level binary optimization.
Discussion
20 comments analyzed.
Concerns raised: No source code published, only JPGs, readme, and Google Drive link - unverifiable claims, VirusTotal flagged file as malicious (1/69 security vendors), Suspicious TCP/UDP traffic on non-standard ports and unknown server IP, Untrusted Windows executable download barrier without video demonstration, Downloaded from unauthorized source without official store distribution
Feature requests: Publish source code to demonstrate skills and build trust, Create video demonstrating the project functionality, Distribute through official storefronts instead of direct downloads
Competitors
Other products that read as similar to this one : 2449 launches clear the similarity bar, closest 8 shown.
Attention rank: #980 of 2450 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 33 days after the earliest competitor.
- CrychicUI · github · 2026-10-19 · 12 upvotes · similarity 0.57
- nanoui · github · 2026-10-18 · 407 upvotes · similarity 0.55
- framelane · github · 2026-10-20 · 15 upvotes · similarity 0.54
- I built a museum of AI interfaces · hn · 2026-08-24 · 6 upvotes · similarity 0.54
- MujicaUI · github · 2026-10-05 · 24 upvotes · similarity 0.53
- rune · github · 2026-10-22 · 78 upvotes · similarity 0.53
- hobby-os · github · 2026-10-13 · 14 upvotes · similarity 0.52
- I run a vision model on every screenshot, locally, on a 4GB GPU · hn · 2026-06-13 · 36 upvotes · similarity 0.52
The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.