How I Launched 2 Profitable SaaS Products in 3 Weeks
I built resumefast.io in 7 days and feedhog.com in 14. Both made money in week one. Here's exactly how I did it, what I skipped, and why boilerplates changed everything.
I've built 4 profitable SaaS products now. The first two took months each. The last two? Three weeks combined.
ResumeFast -- an AI-powered resume builder -- took 7 days. FeedHog -- a customer feedback collection tool -- took 14. Both generated revenue in their first week live.
This isnt a flex post. Its a breakdown of exactly how I did it, what shortcuts I took, and what I'd tell you if you wanted to do the same thing.
The Backstory: Why Two Products at Once
I've written before about the true cost of building SaaS from scratch. The short version: I calculated 250+ hours of infrastructure work before writing a single feature. Auth, payments, email, analytics -- the boring stuff that eats your first 2-3 months.
After building Vibestacks -- the boilerplate I wished existed when I was starting out -- I wanted to prove it actually works. Not with a demo app. With real products that real people pay for.
So I decided to build two back-to-back.
The idea was simple: if the boilerplate is as good as I think it is, I should be able to go from zero to paying customers in days, not months. Two products would prove it wasnt a fluke.
Product 1: ResumeFast -- 7 Days From Zero to Revenue
ResumeFast is an AI-powered resume builder. You paste your experience, pick a template, and the AI rewrites your bullet points to actually sound impressive. Then you download a clean PDF.
I chose this because the market is massive and the existing tools are either bloated enterprise software or free tools with terrible output. Theres a clear gap for something fast, clean, and AI-native.
The Day-by-Day
Days 1-2: Cloned Vibestacks, configured the project, designed the core resume editor UI. Auth, payments, email, analytics -- all working out of the box. I spent zero time on any of that.
Days 3-4: Built the AI resume generation pipeline using the Vercel AI SDK (already configured in the boilerplate). Prompt engineering for different industries and experience levels. Template system for PDF output.
Days 5-6: Polish. Mobile responsiveness, edge cases in PDF generation, pricing page, landing page copy. Tested the full flow probably 50 times.
Day 7: Launched on socials. First paying customer within hours.
What Vibestacks Handled (So I Didnt Have To)
| Component | Estimated Hours Saved | What I Would've Built |
|---|---|---|
| Authentication | ~30 hrs | OAuth, email/password, session management, password reset |
| Stripe Payments | ~40 hrs | Checkout, webhooks, subscription management, customer portal |
| Email System | ~10 hrs | Resend setup, templates, deliverability config |
| Analytics | ~10 hrs | PostHog setup, event tracking, user identification |
| SEO & Blog | ~15 hrs | Meta tags, OG images, sitemap, blog infrastructure |
| Security | ~15 hrs | Rate limiting, CSRF, bot protection, security headers |
| Total | ~120 hrs | 3+ weeks of full-time work |
Those 120 hours turned into roughly zero. I literally just changed environment variables and started building ResumeFast features on day one.
Product 2: FeedHog -- 14 Days, More Ambitious
FeedHog is a customer feedback collection tool. Think of it as the simplest possible way to collect, organize, and prioritize feedback from your users. Public voting boards, embeddable widgets, automatic categorization.
This one took two weeks instead of one because the domain is more complex. A resume builder is mostly a single-player experience -- you make a resume, you download it. A feedback tool is multiplayer. You need:
- Public boards where users can submit and vote on ideas
- An embeddable widget that works on any site
- Admin dashboards for sorting, tagging, and prioritizing feedback
- Notifications when users vote on or comment on your ideas
- API access so teams can integrate with their existing tools
More surface area means more time. But the infrastructure was still instant.
The Meta Part
Heres the fun part: im using FeedHog to collect feedback on Vibestacks itself. Dogfooding at its finest. When a Vibestacks user wants a feature, they submit it through FeedHog. Other users vote on it. I build whats most requested.
It closes the loop perfectly. One product improving the other.
What Made FeedHog Take Longer
The extra week wasnt infrastructure -- that was still day-one stuff thanks to the boilerplate. The extra time went to:
- Designing the widget embed system (had to work across different frameworks)
- Building the voting and ranking algorithm
- Public board permissions (some feedback is internal-only)
- Real-time updates when votes come in
- FeedHog's API for third-party integrations
All product features. All the stuff that actually differentiates the product. Zero time wasted on "how do I set up Stripe webhooks again?"
The Numbers (Honest Breakdown)
| Metric | ResumeFast | FeedHog |
|---|---|---|
| Days to build | 7 | 14 |
| Days to first revenue | 1 (launch day) | 3 |
| Infrastructure time | ~0 hrs | ~0 hrs |
| Feature development time | ~50 hrs | ~100 hrs |
| Lines of boilerplate code used | ~15,000 | ~15,000 |
| Lines of custom code written | ~3,000 | ~6,500 |
Compare this to my earlier products where I spent 250+ hours just on infrastructure before writing a single feature. The difference is night and day.
If we're being honest about the economics: those first two products probably cost me $20,000+ in opportunity cost (at a modest $75/hour). ResumeFast and FeedHog combined cost me about $10,000 -- and 100% of that went into actual product development.
What I Actually Spent Time On vs What I Skipped
This is the part that surprises people.
With my first SaaS products:
- 90% infrastructure, 10% product features
- Weeks before I could show anyone anything
- Burned out before I even launched
With ResumeFast and FeedHog:
- 10% configuration, 90% product features
- Had something to show on day 2
- Launched while I still had momentum
The ratio completely flipped. And that changes everything about how it feels to build. When youre making visible progress every day -- not debugging webhook handlers -- you actually want to keep going.
5 Lessons From Launching Two Products in 3 Weeks
1. Infrastructure Is the Tax on Creativity
Every hour configuring auth flows is an hour you're not spending on the thing that makes your product different. I used to think building infrastructure was "real engineering." Now I think its a tax. Minimize it.
2. First Customers Dont Care About Your Code Quality
Nobody who signed up for ResumeFast asked what ORM I use. Nobody who uses FeedHog cares whether my auth is custom-built or from a boilerplate. They care if the product solves their problem.
Ship quality you're not embarrassed by. But dont let perfect be the enemy of profitable.
3. Momentum Compounds
Building the second product was faster than the first, even though it was more complex. Why? Because I was already in the zone. I knew the stack, I knew the patterns, I had muscle memory for the common operations.
If you have two ideas, build them back-to-back. The momentum from the first carries into the second.
4. Charge From Day One
Both products launched with paid plans on day one. No free beta. No "we'll figure out pricing later." You know what happened? People paid. Immediately.
If your product solves a real problem, people will pay for it. Delaying pricing just delays validation.
5. Boilerplates Are Leverage, Not Cheating
I used to feel guilty about using boilerplates. Like I wasnt a "real" developer if I didn't build everything from scratch.
Thats like saying a carpenter isnt "real" because they buy lumber instead of growing trees. The craft is in what you build, not in how you source your materials.
The Tech Stack
Both products are built on the same foundation:
- Framework: Next.js 15 (App Router)
- Auth: BetterAuth (email/password, OAuth, magic links)
- Payments: Stripe (subscriptions, one-time, usage-based)
- Database: Drizzle ORM + PostgreSQL
- AI: Vercel AI SDK with multiple providers
- Email: Resend with React Email templates
- Analytics: PostHog (events, funnels, feature flags)
- Hosting: Vercel
- Security: Cloudflare (DDoS, bot protection, WAF)
All of this came pre-configured from Vibestacks. I just added the product-specific code on top.
Whats Next
ResumeFast is getting interview prep features -- AI mock interviews and answer coaching. Resume building is just the entry point.
FeedHog is expanding into roadmap management and changelog generation. Feedback in, public roadmap out, changelog when you ship.
And im already thinking about product number 5. When each product takes 1-2 weeks instead of 3-6 months, you can afford to experiment. Some will work, some wont. But the cost of trying is so low that not trying is the bigger risk.
I also wrote about why vibe coding is killing the 6-month MVP and how this trend is reshaping the entire indie SaaS landscape. If youre thinking about building, the barrier has never been lower.
If you want to build like this, check out Vibestacks. Its the exact boilerplate I use for every new product.
Questions? Hit me up at raman@vibestacks.io, on LinkedIn, or in the Discord.
Read more
Vibe Coding Is Killing the 6-Month MVP -- And That's a Good Thing
The era of spending months on infrastructure before writing a single feature is over. Here's how vibe coding, boilerplates, and AI are reshaping how indie hackers build SaaS.
The True Cost of Building SaaS Infrastructure From Scratch
I calculated every hour spent on auth, payments, emails, and analytics. The number surprised me. Here's the real cost of 'just building it yourself'.
Why I Spent a Week Studying API Design Before Writing Code for vibestacks
Building a SaaS boilerplate means other developers will build on top of your API. I had to get this right. Heres what I learned.