I build AI-native engineering software for mechanical engineers: CAD and drawing review, CAE workflow automation, DFM reasoning, and desktop tools that let AI agents work inside real engineering environments.
My current work is centered on a simple belief: engineering software should understand geometry, drawings, requirements, manufacturing constraints, and the context of the engineer using it. The best AI tools in this space will not be generic chat windows. They will be traceable, visual, tool-aware systems that sit close to CAD, simulation, PLM, and review workflows.
I am building toward a stack of engineering assistants that can inspect technical drawings, reason about manufacturability, connect to CAD/PLM workflows, and eventually help generate or repair engineering artifacts with human review in the loop.
The near-term focus is RapidDraft: AI drawing review for mechanical engineering teams. The broader vision is a "Cursor for hardware engineers" where CAD, CAE, requirements, manufacturing rules, and product knowledge are available to the engineer through a coherent AI-first workspace.
- CAD and mechanical engineering workflows: NX, Teamcenter-style review loops, drawing quality checks, tolerance review, DFM, and CAE post-processing.
- AI systems for engineering: retrieval, multimodal review, structured reasoning, agentic tool use, local evaluation loops, and human-in-the-loop validation.
- Product engineering: React, TypeScript, Electron, Vite, Python services, local sidecars, MCP servers, browser automation, and deployment to Vercel, Netlify, and Railway.
- Desktop integration: secure Electron embedding, local capture utilities, app/window sharing, managed browser sessions, and workflow dashboards.
Some of my strongest recent work is product or infrastructure code that is intentionally private. Where a repo is public, I link it. Where the repo is private, I link the product or describe the project without exposing a dead private URL.
| Project | What it is | Stack / focus | Link |
|---|---|---|---|
| RapidDraft | AI drawing review for mechanical engineering teams. Current path: review first, then drawing generation and cost estimation. | CAD drawings, DFM, product workflows, TypeScript, Python | rapiddraft.ai |
| RapidShot | Desktop utility that lets Codex or Claude inspect a shared screen, app window, or managed browser tab through a local MCP server. | Electron, Node.js, MCP, desktop capture, local agent tooling | Private repo |
| Workspace Dashboard | Windows-focused Electron workspace dashboard that embeds trusted web apps as movable tiles with persistent workspaces. | Electron 40, React, TypeScript, WebContentsView, hardened app embedding | repo |
| OneNote Link v4 | Chrome extension that exports OneNote pages through Microsoft Graph and OneDrive, then creates shareable links. | Chrome MV3, Microsoft Graph, OAuth, content scripts | repo |
| QuickChecklists | Simple checklist generator inspired by practical engineering and review workflows. | TypeScript, web UX, lightweight product prototyping | repo / demo |
| Research Portfolio | Public research portfolio site for presenting technical work and research narratives. | HTML, static web, portfolio design | repo / demo |
I like tools that make technical judgment easier to apply: systems that reduce review friction, preserve engineering context, and make hidden assumptions visible. The projects above are all versions of that same theme, from CAD drawing review to local agent utilities to dashboards for focused technical work.
If you are working on engineering AI, CAD automation, DFM, CAE, or serious desktop tooling, I would be happy to connect.
Email: adeelyj@gmail.com