Building Vuti — i built it because i was posting for a mobile toilet business

i built vuti because i was posting for a mobile toilet hire business.

no, seriously.

the side hustle

i'm a software engineer. but i also run a side hustle — mahapa event hire. mobile toilets. fridges. events.

picture this: i'm on-site fixing a VIP toilet, rubber gloves on, and i still need to post consistently for the business. weddings on saturday, corporate events the next weekend, the calendar doesn't care that my hands are full.

so i tried to get help.

the help that didn't exist

i couldn't find the right person. someone who would:

• stay consistent
• understand the brand
• not need babysitting

freelancers ghosted. agencies were either too expensive or too generic — they'd post for a "service business" without ever capturing what mahapa event hire actually sounds like. AI tools could write something, but the something was off. wrong tone, wrong angle, sometimes plain weird.

so i built the tool myself.

what i built

vuti plans, drafts, and schedules — you just approve before anything goes live.

it learns your voice from your existing content. it plans a calendar. it drafts copy and images for each platform. then it queues posts up for you to review. you approve in minutes, vuti publishes to LinkedIn, X, Bluesky, and Facebook.

the workflow is the whole thing:

plan → review → scheduled → published

every post moves through that pipeline. you always know what's coming, what needs your eyes, and what's already live.

then it hit me

i was building this for myself. but it wasn't just me.

the plumber. the caterer. the fridge hire guy. the small shop owner. the wedding photographer. the personal trainer.

they all have this exact problem. they're great at the thing they do. they know they should be posting. they don't have time, don't have a brand voice doc, and don't want to hire someone they'll have to manage.

that's who vuti.io is for.

the build

the stack stayed boring on purpose. i wanted to ship, not architect.

• frontend: next.js, typescript, tailwind
• backend: node.js for the api, python for content generation
• AI: OpenAI for copy and image generation
• integrations: linkedin, x, bluesky, facebook

a lot of the work isn't AI. it's the un-glamorous stuff: a calendar that shows the right thing on the right day, an asset library that's actually findable, previews that match each network, a catalog of products that vuti can pull into posts.

the AI is the engine. the workflow is the car.

what i've learned so far

a few things from building this:

1. build for one user first. i was the one user. i knew exactly what i needed because i was the one cancelling posts at 9pm because i'd been on-site all day. when you build for one real person you trust, you don't have to guess.

2. the right customer is already paying for the problem. small business owners aren't asking "should i post on social?" — they're already losing weekends to it, or paying agencies, or just not posting and feeling guilty. you don't have to convince them the problem exists. you just have to be better than what they're doing now.

3. approval is a feature, not a friction. people don't want an AI yeeting random posts onto their feed. they want help getting from blank page to "yeah, send it." vuti is built around that moment.

4. ship the thing you'd use at 11pm with rubber gloves still on. literally my bar.

where it's going

the beta is live. LinkedIn, X, Bluesky, Facebook. more networks coming. sharper analytics. better brand-voice modelling so month two feels noticeably better than month one.

if you're a small business owner reading this and you're tired of staring at a blank post composer, try vuti.io.

and if you're ever throwing an event in polokwane and need mobile toilets, fridges, a tent, or chairs — mahapa event hire has you covered.