Josh001 - Stardance

Ship Changes requested

@Josh001 on Dashboard Roadmap Generator · 8 days ago

WHAT I BUILT: A modern browser dashboard roadmap generator that updates the UI dynamically using a clean client-side JavaScript fetch API.

WHAT WAS CHALLENGING: Overhauling the traditional HTML form reloads into asynchronous JSON fetch streams, and configuring custom serverless path routing to ensure Flask could run natively inside Vercel's hosting environment.

WHAT I AM PROUD OF: Transitioning the application into a highly professional, modular frontend/backend split architecture without relying on low-code frameworks or AI website builders.

WHAT PEOPLE SHOULD KNOW: Enter any age (between 0-120), your skill tier, and core focus area into the dashboard inputs. Hit the generate button to instantly view a structured 30-day, 6-month, and 12-month study roadmap dynamically populated on your screen with zero page reloads.

Try project → See source code →

Open comments for this post

@Josh001 on Dashboard Roadmap Generator · 8 days ago

33m 59s logged

Overhauled the existing repository structure specifically for the Stardance mission rules. I split the codebase into clean frontend and backend folders, replaced the traditional HTML form reloads with an asynchronous JavaScript fetch API system, and successfully deployed the final dynamic environment to Vercel with serverless path routing configurations.

here s mine deploy link : https://roadmap-generator-dun.vercel.app/
here s mine github link : https://github.com/williamjosh-dev/roadmap-generator

Open comments for this post

@Josh001 on physical sandbox v1.0 · 26 days ago

53m 32s logged

4D AI Sandbox Engine 🪐

hey everyone, making my first post here.

i’ve been working on a 4d ai sandbox prototype that takes a prompt, uses an llm to compile glsl code on the fly, and injects it straight into a live webgl runtime loop.

What is running under the hood?

GPGPU pipeline: all particle calculations are offloaded entirely to custom fragment shaders using texture buffers so the cpu doesn’t choke.
Live rendering: currently pushing a 128x128 grid (16,384 nodes) completely smooth on localhost.
Terminal stream: built a custom execution stream log on the bottom left to track engine boots and live mapping errors.

What is the plan for v1.0?

I want to turn this prototype into a practical tester for 3d assets. Next steps:

Map the gpgpu coordinates into actual solid surfaces using three.js InstancedMesh.
Build an exporter node so users can download the generated shapes as standard .gltf files.
Finish the sidebar chart to track structural/physics stress metrics over time.

P.S. — attached a screenshot of how the dashboard looks right now. let me know if anyone wants to geek out over gpu textures or shader pipelines!