Ozone Generator / Fab2 Internship
May 2026 - August 2026
Primary tool owner for two generations of semiconductor ozone generators, deployed in SF and ATX; +83% uptime, +230% weekly output.
tldr
- Role
- Tool owner, semiconductor ozone generators · Vacuum Systems intern · Fab2, May–Aug 2026
- Built
- Root-caused a tolerancing (GD&T) failure and a thermal failure across the generator fleet; productionized gen 2 at ~30% the size; designed gen 3
- Result
- Uptime +83%, weekly ozone output +230%, power per % concentration −47%; gen 3: 58% fewer unique parts, ~85% lower cost
- Tools
- GD&T, high-voltage transformers, gas handling, thermal management, procurement
Fab2 (formerly Atomic Semi) was an amazing internship experience. I spent ~100 days on the vacuum systems team, and spent the entire period working on one project.
I spent my internship improving a fleet of generators designed before I arrived, designing + productionizing a second generation to reduce parts/costs, and designing a considerably better machine to be produced in future fabs. I got to service machines while lying on the floor of Fab2's fabs, design experiments with thousands of dollars of components, and ask tons of questions while building BOMs with procurement and even creating recipes with the process team.
Through remedying a GD&T problem and a thermal problem in the original fleet, I increased machine uptime by 83%, raised average weekly ozone output by mass by 230%, and reduced power per percent of ozone concentration by 47%. My second-gen machine was ~30% the effective size of the original, and my third generation design had 58% fewer unique components and a ~85% reduction in price if compared directly to the original solution.
To shed more light on the ownership I was given at Fab2, I remember being told in my second week that I had a delivery date in less than 14 days. This date had been set a few weeks before I arrived, and was a requirement for an ozone generator needed for another project that other engineers were relying on. To meet it, I had to disassemble and learn the existing machines, fix two historically underperforming ones, and then schedule time with the project stakeholders to pass acceptance tests and hand them off. It was a blast.
I can't share much about the technical side of my work (the image on this page was the only one I was allowed to ever download), but my role allowed me to learn a ton about high voltage safety, gas handling, dielectric materials, transformer design, and thermal management. I engaged heavily with our supply chain while procuring parts for prototyping and a production build of the second-generation generator (pictured on the right, with everything exciting hidden inside and out of view).
