When a one-hour tool beats a monthly subscription

Project Link : https://staff-creator--pratikjadhav41.replit.app/

The task nobody had time for

AdvaRisk's HR team had been sending monthly newsletters through a third-party tool. New joiners, exits, work anniversaries, team wins. It worked fine until it went paid. Rather than subscribe, the company decided to build something in-house, but the tech team had bigger things to deal with. So it landed with me. My first move was the obvious one. I designed the newsletter template in Figma, exported it as an image, and emailed it out. On-brand and done. I ran that for three or four months, and it worked, but it didn't scale. Every month meant reopening the file, editing names, swapping photos, nudging layouts. Tedious doesn't quite cover it.

Testing vibe coding on something real

I'd been hearing a lot about vibe coding, the idea of building working tools by iterating with AI rather than writing everything from scratch, and I wanted to test it properly instead of taking the hype at face value. So I gave myself real constraints. A genuine use case, the free plan on Replit, and a one-hour limit. The goal was a form-based tool where anyone on the HR team could enter the month's data and get back a rendered, shareable newsletter image. Within the hour I had it working. It took structured data through a simple form, generated a layout that matched the Figma design, and exported it as an image ready to send. The output landed at roughly 70 to 80 percent of the original Figma, holding the branding, section structure, typography hierarchy, and the yellow-on-navy scheme.

What it proved, and where it falls short

This was never about replacing designers or developers. It was about knowing when a lightweight, fast solution is the right call, and a polished internal tool built in an hour on a free plan that solved a real recurring annoyance is exactly that kind of scrappy problem-solving a growing company runs on. It also settled something I wanted to test for myself. Vibe coding with the right tools is a real prototyping strategy, not a novelty, and the gap between an idea and a working product is now measured in hours, not sprints. The honest caveat is that it was built on Replit's free plan, so the app gets unpublished after 30 days. That's a fair trade for a one-hour experiment, and if the team wants it permanent, moving it to a paid tier or another host is a small step, not a rebuild.