Before Ren, I worked on product teams at two security-minded startups. The work I’m proudest of is noticing something was wrong before a customer did, and getting it fixed while it was still small.
Everything below is redrawn generically. None of it is from either company’s product.
Problem
A slow log-out workaround, client data kept apart only by the screen, and scans that failed without an error.
Solution
An in-app account switcher, a server-side check on every request, and a completeness check right after scanning.
Results
Both Quantech fixes shipped before launch, and my switcher became Spin.AI’s fix.
Switching client accounts meant logging out
Spin.AI protects Google Workspace and Microsoft 365 from data loss. Solutions engineers had to log out and back in to switch between client accounts. I confirmed with customer success that it was a real, repeated pain, then designed the in-app switcher that became the fix. Try both:
Get from Client A to Client B.
The rest of my time at Spin was bigger, slower work. I defined requirements for a partner controls feature, mapped how a partner’s configuration would interact with platform settings, and wrote the BRD engineering built from. I also analyzed more than 20 SaaS products and built over 100 security control mappings that fed the roadmap and the controls library partners use.
Client data was only kept apart by the screen
Quantech builds tax software for accounting firms that handle sensitive client data. Before launch, I noticed firms were only kept apart by the screen, not by the system behind it. I proved it through the API, documented it, and engineering fixed it before launch. Try both routes:
You’re signed in as Firm A. Try to open Firm B’s client records two ways.
Pick a route.
A scan that failed without saying so
In the logs, I found scans that came back incomplete without any error, so bad data looked like good data. I specified a check right after scanning that tells the user immediately.
Incomplete flag to the user
