ForgeIQ is a multimodal AI engineering intelligence platform that converts engineering artifacts (photographs of mechanical components, CAD screenshots, engineering sketches, technical drawings, datasheets, or damaged parts) into structured, professional engineering reviews.
Engineers, product designers, robotics teams, and manufacturing consultants often need rapid initial reviews of mechanical components for DFM (Design for Manufacturability), materials suitability, risk & failure analysis, and quality inspection. Manual early-stage engineering audits can take days. ForgeIQ leverages Gemini multimodal intelligence to analyze visual and textual artifact evidence alongside operating context, providing structured, actionable engineering reports in seconds.
- Multimodal Artifact Analysis: Accepts component photos, CAD screenshots, engineering sketches, technical drawings, and technical PDFs.
- 7 Tailored Analysis Modes:
- General Engineering Review
- Design for Manufacturability (DFM)
- Materials and Process Selection
- Failure and Risk Analysis (FMEA)
- Quality Inspection Checklist
- Robotics and Mechanism Review
- Medical Device and Product Review
- Structured Engineering Report:
- Executive Summary
- Artifact Identification & Confidence Level
- Functional Purpose & Interacting Interfaces
- Geometry & Stress Concentration Observations
- Material Assessment & Candidate Alloy/Polymer Tradeoffs
- Manufacturing & DFM Optimization
- Failure Risk Analysis & Qualitative FMEA Table
- Quality Inspection Checklist with Interactive Checkboxes
- Prioritized Engineering Recommendations (Critical/High/Medium/Low)
- Missing Data Checklist
- Contextual Q&A Follow-up Assistant: Conversational engineering assistant grounded in the uploaded artifact and generated report context. Distinguishes hard facts, assumptions, and recommendations.
- Report Export: Copy formatted report text or download as clean Markdown file.
- Built-in Demo Samples: Pre-packaged engineering artifacts (UAV Motor Mount Node, Orthopedic Bone Plate, Robotic Actuator Housing) for immediate live or preset evaluation.
- Frontend: React 18 / 19, TypeScript, Tailwind CSS v4, Lucide React icons
- Backend Runtime: Node.js, Express, ESBuild, TSX
- AI Engine:
@google/genaiTypeScript SDK (Server-side Gemini 3.6 Flash model with structured JSON Schema)
- Node.js 20+
- npm
Create a .env file in the project root:
# Required for Gemini API calls
GEMINI_API_KEY="your_gemini_api_key_here"
# App URL
APP_URL="http://localhost:3000"In Google AI Studio, GEMINI_API_KEY is automatically injected into the server runtime via the Settings > Secrets panel.
# Install dependencies
npm install
# Start development server
npm run devThe application will be accessible at http://localhost:3000.
- Preliminary Analysis Only: ForgeIQ output represents preliminary AI-assisted observations and must be validated by a licensed professional engineer using full engineering drawings, physical testing, simulations (FEA/CFD), and domain standards.
- No Fabricated Data: ForgeIQ explicitly refuses to fabricate dimensions, exact numerical tolerances, or regulatory approvals when evidence is insufficient.
- Medical Devices: Output is framed strictly as early-stage engineering guidance and does not claim FDA/CE regulatory approval or diagnostic validity.
- STEP / IGES CAD file 3D mesh rendering and stress heatmap visualization.
- Direct export to PDF and formal PDF/A technical submission packages.
- Integration with material databases (e.g., MatWeb, Ansys Granta).
- Automated FEA meshing script generation for Abaqus or ANSYS.
To export or push this project to GitHub:
- Initialize git:
git init - Add remote repository:
git remote add origin <your-github-repo-url> - Commit and push:
git add . git commit -m "Initial commit of ForgeIQ application" git branch -M main git push -u origin main