I Have Been Using AI For A Creative Maker Project

TL;DR

Before you orderOffer from Amazon

Get home appliances delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

A maker is using Google Gemini to generate OpenSCAD code for parts of a Fallout-inspired Halloween prop he is building with his son. The AI-generated models have needed repeated prompting and manual edits, but have helped him explore shapes and move toward test prints. The finished prop and its final design are not yet complete.

A maker is using Google Gemini to generate OpenSCAD code for 3D-printed parts of a Fallout-inspired Halloween prop he is developing with his son. The project is still in progress, but the maker says AI has helped him explore designs he found difficult to model from scratch, even though its outputs have required repeated corrections.

The project began after the maker’s son chose a Fallout 4-themed costume. A full DIY power-armor costume was judged impractical within the available time, so the maker shifted toward a simpler costume and a more prominent custom prop. A planned motorcycle chest protector did not arrive as expected, leaving foam and craft materials as possible costume options.

For the prop, the maker used Gemini to produce OpenSCAD code for a shape made from two connected truncated cones. The parametric code allowed him to adjust dimensions before printing. He reports that an initial print, made with wet filament, was useful for moving to the next stage, and that a later version may include holes for attaching the part to a wooden box.

The AI has not produced a finished, reliable design without intervention. In one example, it ignored instructions about keeping side arms flush and limiting pill-shaped holes to the main face, and added chamfers the maker did not want. He says he changed code parameters himself and is testing a low-density print of another part. He may remodel some components in Fusion or similar software for more direct control.

AI Supports Iterative Prop Design

The project offers a practical example of AI-assisted making: a text-based tool can help turn an early idea into editable code and a physical test part, without replacing the maker’s judgment. That can be useful when a project involves unfamiliar shapes or when a person has trouble translating a mental image into a model.

The account also shows the limits of that assistance. The maker still needs to identify errors, revise the code, make choices about assembly and test printed parts. For readers considering AI in a workshop, the report describes it as an aid to prototyping and visualization, not an automatic route to a finished design.

From Costume Idea to Test Prints

The maker says he is comfortable modeling objects he has made before, but finds it harder to design new forms that he has not fully visualized. He turned to Gemini after deciding that a prop needed to stand out alongside a less elaborate costume. His first modeled piece was a truncated-cone assembly, and he later sought help breaking a more elaborate fictional tool into printable subassemblies.

That breakdown was not useful as a complete design, according to the maker, though it supported his earlier thought of using mounting pins for glued assembly. His current plan is less integrated: attach individually designed parts to a small wooden box, with screws, glue or other hardware as needed. He also considered printed air fittings as a way to check how parts and quarter-inch-inner-diameter hoses might fit, while still deciding whether to use inexpensive metal fittings.

“I have been focusing my attention on the props.”

— The maker, in the ToolGuyd report

The Prop Is Still Unfinished

The final prop’s appearance, function and assembly method have not been settled. The maker has floated several possible identities for it, including a detector, and says his son may choose what to call it after they build it. The source report does not establish whether the current test prints will fit the final assembly or whether the project will be completed by Halloween.

It is also unclear which parts will remain AI-generated and which will be remodeled manually. The maker says the first AI-generated subassembly breakdown was not very helpful, and he may move some work to conventional 3D software. The reported experience is one maker’s account; it does not establish how Gemini or other AI tools perform across different design tasks.

A Rough Version Before Halloween

The maker’s stated near-term goal is a rough first-version prop that can be assembled in time, followed by later refinements if there is enough time. He is testing a low-density print and expects to revise dimensions or remake parts before choosing final materials and colors. He may use Fusion or similar software to improve control over the handle and the relationships between components.

For now, the plan is to build the prop with his son around a small wooden box, using printed components and whatever fastening methods prove workable. The report does not give a firm completion date or confirm that the design will reach a polished final form.

Key Questions

What is the maker building?

He and his son are developing a Fallout-inspired Halloween prop to accompany a costume. Its final identity and appearance are not settled.

How is AI being used in the project?

The maker used Google Gemini to generate OpenSCAD code for shapes and components, then adjusted parameters and tested printed parts.

Did the AI produce a finished design?

No. The maker reports that it ignored some instructions and added unwanted features. He corrected code and is considering manual modeling for better control.

Is the prop complete?

No. The project is still at the prototype stage, with test prints and assembly plans in progress. The report does not confirm whether it will be finished by Halloween.

Source: rss