All projects
AI AgentsIn progress

JARVIS

A conversational CAD agent that connects natural-language requests to modeling tools, geometry validation, and an interactive 3D view.

LangChain
Tool Calling
CAD
3D Visualization

Yong-Hwan Lee

Period · Mar. 2026 – Present

Select a bracket face, request two reinforcing ribs, and inspect the updated geometry. This 10-second local demo uses a scripted command with real CAD tool execution.

Overview

JARVIS is a web application for creating and modifying CAD models through conversation. It connects a LangChain agent to modeling tools so that a user can describe a change, inspect the resulting geometry in a 3D view, and continue refining the model.

The project covers the agent workflow, CAD tool integration, validation, revision history, and the web interface. Development began in March 2026 and is ongoing.

Technical contributions

  • Agent and tool integration: Implemented a LangChain agent with structured tool inputs and outputs for inspecting and modifying CAD models. Connected conversational requests to executable modeling operations.
  • Geometry validation: Integrated geometry and dimensional checks, including regeneration from the saved model definition. Validation results help distinguish a generated candidate from a verified model.
  • State and revision management: Preserved model revisions and conversation context so that subsequent requests can refer to an earlier result. Kept candidate changes separate from the accepted model state.
  • Interactive 3D interface: Implemented web-based visualization and selection of model features to connect conversational requests with the geometry being edited.

Implementation

Python manages the agent, model state, validation, and application API. Replicad and OpenCascade perform geometry operations in an isolated server-side Node.js process. The browser uses Three.js to display the resulting mesh and support interaction with the model.

The workflow connects conversation → tool execution → validation → 3D review, with revision history available for subsequent edits.

Current scope

The implementation supports a defined set of modeling operations and validation checks. It is under active development, with a focus on connecting agent decisions to inspectable CAD results.

Keep exploring

Civil & Construction AI Agent