Two Berkeley Grads Left AWS to Build the World's First AI Mechanical Engineer

The Corporate Exit

Arya Bastani was a cloud architect at AWS. Keaton Elvins helped launch Amazon Q. Roth Vann was building at Meta. Three Berkeley engineers with the kind of resumes that guarantee six-figure jobs for life.

They quit anyway.

The insight was specific: mechanical engineers spend 60-80% of their time on repetitive CAD work that follows predictable patterns. Design a bracket, run the stress analysis, iterate on tolerances, generate manufacturing drawings. The creative work - the actual engineering judgment - is maybe 20% of the job.

What Camfer Actually Does

Camfer built an AI mechanical engineer. You describe what you need in natural language - "design a mounting bracket for a 50kg load with M8 bolt holes" - and it generates parametric CAD models, complete with manufacturing drawings.

This is not a chatbot that suggests ideas. This is an autonomous agent that does the work.

Why This Matters

The mechanical engineering market is massive. Every physical product that gets manufactured needs engineering drawings. Every factory, every hardware startup, every automotive supplier.

The old model: hire a team of 5-10 mechanical engineers at $80-120K each. Wait weeks for design iterations. Pay for expensive CAD licenses.

The Camfer model: describe what you need. Get it in minutes.

When three engineers from AWS and Meta leave to build the same thing, pay attention. They saw the inside of big tech and decided building the replacement was better than being the replaced.

The Breakout Pattern

Camfer follows a pattern we see again and again: engineers who worked inside the machine, understood what's automatable, and left to build the automation.

They didn't raise $100M. They didn't hire 50 people. Four engineers, $4.8M, and a thesis that AI can do what mechanical engineers do 80% of the time.

That's the breakout.