Artifact is a generative AI platform that integrates physics-based reasoning with open-source engineering tools to streamline systems engineering, modeling, and simulation tasks. It enables engineers to efficiently create simulations, design control systems, and optimize system-level designs while maintaining data integrity and collaboration across projects.
Funding
Funding not disclosed
Founders
Product
Problem
Systems engineering, modeling, and simulation often involve complex toolchains, manual coding, and ensuring data integrity across various software tools and model components. This complexity can hinder efficient design iterations and collaboration, especially when integrating physics-based reasoning.
Solution
Artifact is an AI-native platform designed to streamline physics-based engineering workflows by integrating AI-driven physics reasoning with open-source engineering tools. The platform enables engineers to efficiently create simulations, design control systems, and optimize system-level designs. Artifact's semantic layer enforces physics principles, automates the assembly of analysis artifacts, and manages interfaces between software tools, data files, and model components. By maintaining a single-source-of-truth system specification, Artifact facilitates collaborative workflows and ensures data integrity across projects.
Target Audience
Artifact is primarily targeted towards systems engineers, modeling and simulation specialists, and control systems designers working on physics-based engineering projects.
Features
- AI-driven physics reasoning engine that understands the physical world
- Automated generation of 3D models in Engineering Sketchpad (ESP) format for packaging and configuration studies
- Aerodynamics model generation using USAF Digital Datcom, parsed into Python code
- Integration with the Python Control Systems Library for dynamics and controls problems
- Collaborative workflows with a single-source-of-truth system specification
- Specialized AI agents for tasks like system identification, software-in-the-loop testing, and Monte Carlo studies
- Ability to run on-premise and integrate with proprietary, air-gapped company toolsets and data