Viclair

29s nachgedacht Inspiration

I am a master plasterer, not a traditional software developer. I´m building a self transforming app with a self refractoring ai Architekture but i couldnt finish that app in time . also Five days before this submission, I started building VicLair because I saw a gap in photo editing: professional tools are powerful but often overwhelming, while simple apps hide important decisions behind filters and one-click effects.

I wanted something in between — an editor that feels approachable, but still gives the user real control.

What it does

VicLair turns photo editing into a clear, intelligent workflow. Users can open an image, make controlled adjustments, compare the result with the original, and export the finished photo.

The long-term idea is to combine the accessibility of a mobile photo app with the editing depth of professional software. AI should not replace the user’s taste or secretly alter the image. It should understand the photo, suggest useful improvements, explain its reasoning, and translate those suggestions into visible adjustments that remain editable.

How I built it

VicLair was created from scratch during OpenAI Build Week using Codex and GPT-5.6 as development partners.

Codex helped me build the application structure, implement the editing workflow, connect the interface to the image-processing pipeline, and diagnose technical problems. GPT-5.6 helped refine the product concept, challenge design decisions, structure features, and turn rough ideas into concrete implementation tasks.

The project moved from an idea to an early working application in only five days.

Challenges

The biggest challenge was not generating code. It was making the entire workflow reliable.

Image editing requires the preview, controls, internal state, and exported result to remain consistent. At the same time, the interface must stay simple enough for someone who does not understand professional photography terminology.

Building, testing, debugging, and redesigning these parts within five days forced me to focus on the most important user journey instead of trying to complete every planned feature.

Accomplishments

The most important result is that VicLair became more than a concept. It is now a real early-stage product with a visible editing workflow and a foundation that can be expanded.

For me personally, the project also proved something larger: someone from outside the traditional software industry can use modern AI tools to turn an original product idea into working software.

What I learned

AI can dramatically reduce the distance between an idea and a prototype, but it does not remove the need for judgment, testing, and persistence.

The best results came when I treated Codex and GPT-5.6 not as automatic app generators, but as collaborators that could propose, build, critique, and revise individual parts of the product.

What’s next

The next steps for VicLair are deeper AI-guided image analysis, transparent editing recommendations, local adjustments, reusable editing styles, batch workflows, and a personal photo library.

The goal remains simple:

Professional-looking photo editing without taking control away from the person behind the image.

Built With

• claude
• claudedesign
• codex