The fastest SaaS products we've built didn't use the newest stack. They used the most boring one.

What we reach for when a founder needs a working product:

• Next.js App Router, React, TypeScript and Tailwind

• Supabase Auth and Postgres, Prisma for the schema

• Stripe subscriptions kept in sync with webhooks

• Transactional email for the moments that matter

• A marketing site in Webflow or Framer the team can edit

What we skip at MVP stage: microservices, Kubernetes and anything the first 100 users won't notice.

Every novel choice is one you'll debug at 2 a.m. Save the novelty for the part customers pay for.

Founders: what did you over-build in your first version?

Really agree with keeping the stack simple early on. Curious, how do you decide when a SaaS product has outgrown its MVP architecture? Would love to exchange thoughts on this in DMs.

Another project from the archive: a website refresh for Ecorp Systems. The team had already chosen a Webflow template and mapped out the site structure. My job was to adapt it to their content and build the website. I worked through the layouts in Figma first, turned the template sections into reusable components, and added a few custom sections where needed. The result: 16 pages built in Webflow, with how-to videos to help the team make basic content updates. Here’s a quick look.

The how-to videos for the team are the part most people would skip, and they're probably the most valuable thing in the whole project. A 16 page Webflow build means nothing if nobody dares touch it afterward. Did the team actually start making their own updates, or did they still come back to you?