LLM-based Visual Code Completion for Aerospace Geometric Design

2026-06-15

AI Summary

This study addresses the absence of large language model (LLM)-driven geometric design assistants in aerospace engineering that simultaneously ensure safety and interpretability. The work introduces, for the first time, the ReAct reasoning framework into visual programming within this domain, presenting an LLM-based programming assistant tailored for expert-level geometric modeling. Built upon GPT-5.4 and a vision-language model, the system integrates a custom Grasshopper plugin, Wingbuilder, and a domain-specific dataset—AVPD—comprising 18 representative tasks. User evaluations with two senior engineers indicate that the system’s suggestions offer practical utility, particularly for complex and time-consuming tasks, though inference latency currently limits real-time interactivity. This research establishes a novel, interpretable, and verifiable paradigm for intelligent programming in high-assurance engineering design.

Abstract

Recent advances in both Large Language Models (LLMs) and Vision Language Models (VLMs) have seen a step change in their ability to perform visual code completion, but the aerospace industry, which prioritizes safety and explainabilty over rapid LLM adoption, currently has no publicly announced LLM-based geometric design copilot systems in commercial use by aerospace Original Equipment Manufacturers (OEMs). This paper presents a LLM-based visual programming copilot application for aerospace engineering design tasks, using a visual programming variant of the ReAct methodology and GPT 5.4. In addition to the copilot, we describe Wingbuilder, a new Grasshopper plugin library with custom components for aerospace-specific geometry abstraction, and an associated Aerospace Visual Programming Dataset (AVPD) with 18 aerospace expert designed tasks at different levels of difficulty alongside ground truth solutions. We evaluate our copilot application with a user trial involving two experienced aerospace engineers from a large aircraft manufacturing company. We find our copilot visual programming ReAct methodology was successful in generating suggestions that participants found helpful, but slow ReAct inference times limit its usefulness to more complex time-consuming tasks where waiting for good copilot solution suggestion was worthwhile. Participants reported they liked the tool and would be willing to use it in the future.

Problem

• LLM-based visual code completion
• aerospace geometric design
• visual programming copilot
• safety and explainability
• aerospace OEMs

Innovation

• Visual Programming
• ReAct Methodology
• Aerospace Geometric Design
• LLM-based Copilot
• Domain-specific Dataset