Not the highlight reel — the honest version of building Frittu solo alongside a full-time job, from the idea that almost didn't happen to launch.

I Built a Fridge-Scanning Recipe App as a Solo Developer in Melbourne — Here's the Honest Version

Not the highlight reel. The actual story.

I want to be upfront about something before I start: I almost didn't build this.

Not because I didn't believe in the idea. Because I've believed in a lot of ideas. I've stress-tested concepts for a PT management app, a real estate tool, a restaurant POS automation, a dog grooming SaaS. I have a framework I run every idea through before I write a single line of code — questions designed to kill the idea before I fall in love with it. Most ideas don't survive.

Frittu survived. But only barely, and not for the reasons I expected.

The Problem Was Mine First

I work full time as an Oracle EPM architect. The job is demanding, mentally intensive, and doesn't leave a lot of space for "what's for dinner tonight." I'm also someone who cares about eating well — not obsessively, but intentionally. I have protein goals. I have cuisines I love. I have things I don't eat.

The pattern I kept falling into was this: I'd get home, open the fridge, stare at it, and feel paralysed despite the fact that there was clearly food in there. The disconnect between "I have ingredients" and "I know what to cook" was costing me time every single night. I'd either order in — spending money I didn't need to spend — or I'd cook the same five meals I always cook because they were the only ones I could execute on autopilot.

I tried recipe apps. I tried meal planning apps. The problem with most of them is they're built around an assumption I couldn't reliably meet: that you've shopped intentionally for the week's meals. The whole system starts with a recipe, generates a list, sends you to the store. If your fridge doesn't match the plan — and mine often didn't — the system breaks.

What I actually wanted was the inverse. Tell me what I have. Tell me what to cook. Do it in a way that matches what I'm trying to eat. That app didn't exist, so I built it.

Why I Almost Talked Myself Out of It

The kill critic framework I mentioned isn't just a phrase. I genuinely ran Frittu through hard questions before starting:

Is the market big enough, or is this a personal problem that only feels universal? Is the competition so entrenched that a solo developer can't get traction? Does this require infrastructure I can't build or afford alone? Is the problem real enough that people will pay to solve it?

The competition question was the one that gave me the most pause. There are large, well-funded recipe apps. There are meal planning tools with hundreds of thousands of users. There are AI-powered cooking assistants from companies with teams I'll never match.

What I kept coming back to was the specific problem. Fridge-first. Scan what you have, get what to cook. Most of the big players aren't built around that starting point — they're recipe libraries first, personalisation second, and "what's in your fridge" is usually a search filter, not the foundation of the product.

The niche was real. The frustration was real. I decided to build.

What It Looks Like to Build an App Solo

I want to be specific here because "solo developer builds app" sounds romantic and the reality is more complicated.

My tech stack: Flutter for iOS, Supabase for the backend, AI for vision and recipe generation, RevenueCat for subscriptions, Vercel for the marketing site. All tools I'd worked with before in various combinations, which mattered — a new stack on a solo project is a risk multiplier I didn't need.

The build happened alongside a full-time job. Early mornings, evenings, weekends. The kind of schedule that's sustainable for a few months but not indefinitely, which meant scope control was non-negotiable. Every feature that wasn't core to the fridge scan → recipe generation loop was deferred. A meal scanner feature I'd planned extensively sits in the backlog. It'll ship post-launch. It's not launch.

The hardest part of building solo isn't the code. It's the decision-making without a second opinion. Every UX call, every pricing decision, every copy choice — you make it, you live with it, and you don't have the luxury of a design review or a product discussion. You develop an internal critic and you get faster at arguing both sides of every decision yourself.

The second hardest part is the switching cost between modes. One morning I'm debugging a RevenueCat entitlement issue. That afternoon I'm writing landing page copy. That evening I'm thinking about Instagram content strategy. These are genuinely different mental modes and the transitions are more expensive than they look from the outside.

The Decisions That Actually Mattered

Some technical decisions are reversible. Others calcify early and are expensive to change. The ones I think hardest about on Frittu:

Fridge scan as the primary entry point, not a feature. A lot of apps have a "cook from fridge" mode buried three taps in. I wanted it to be the first thing — the hero action when you open the app. This shaped almost every other UI decision downstream.

Pricing model: free with a meaningful free tier, not a paywall. I've seen apps that gate so aggressively that you can't evaluate whether the product is actually good before paying. The Frittu free tier gives you real scans, real recipes, a real taste of what the app does. The upgrade to Pro is for people who want more — not people who hit a wall after five minutes. I think the industry trend toward hard paywalls is a short-term revenue optimisation that destroys long-term trust.

Orange as the brand. This sounds like a small decision but it defined everything. I tested a linen-and-green palette that looked sophisticated and clean and felt completely wrong for an app about cooking. Orange is warm, energetic, and appetite-adjacent — there's research behind why fast food brands use it, and it's not an accident. The mascot, the onboarding screens, the icon — they all work because the colour decision was made early and held firmly.

Australian-first positioning. Frittu's shopping list integration names Coles, Woolies, Aldi, IGA. The copy is written for an Australian user. "Made in Melbourne" is in the footer. This is deliberate. Most recipe apps are built in the US for a US user and localised (badly) for everywhere else. Being actually local is a product differentiator, not just a marketing line.

What I Got Wrong

Honesty requires including this section.

I underestimated how long App Store submission prep takes the first time. Not the technical side — the metadata, the screenshots, the review guidelines, the privacy disclosures, the app preview video requirements. There's a whole layer of non-code work that first-time App Store developers don't account for and I was no different.

I also initially set the wrong pricing in RevenueCat versus App Store Connect, which created a discrepancy I had to resolve before launch. Small thing, fixable, but the kind of thing that doesn't show up in tutorials and costs you an afternoon when you hit it.

The onboarding flow went through more iterations than any other part of the app. I kept trying to show too much too fast — the full feature set, the personalisation options, the meal planning, the shopping list — and it consistently overwhelmed test users. The version that shipped is the one that restrains itself to four slides and trusts the product to show its value once you're inside it.

What I Think Frittu Actually Is

On the surface it's a recipe app. But I think the real thing it does is remove the most draining micro-decision in most people's days.

"What's for dinner?" is a question that sits in the background all afternoon. It creates low-level cognitive load that most people don't even notice because it's so constant. The decision compounds if you're also trying to eat to a goal, avoid certain ingredients, or not waste the food you already bought.

Frittu doesn't just answer the question. It makes the question unnecessary. Your fridge has food. Your goals are set. Here's what you cook tonight. That's the version I wanted when I started building, and that's the version I shipped.

Why Melbourne Specifically

I want to make a point about this because it's something I think about.

A lot of the indie app building conversation happens in US cities — San Francisco, New York, Austin. The audience, the investors, the press, the communities — they skew heavily American. Building from Melbourne means you're on the other side of the world from most of that, operating on a time zone that makes real-time conversations with US communities genuinely inconvenient.

But Melbourne is a city that eats well and cares about it. It's multicultural in a way that shapes how people cook — the range of cuisines in a typical Melbourne household's rotation is genuinely broader than many cities. It's a city where people go to the Queen Vic Market on Saturday morning and think about what to cook from what they find.

Frittu isn't a Melbourne app in a limiting sense. But it was built by someone who thinks about food the way Melbourne thinks about food — as something worth taking seriously without taking too seriously.

What Comes Next

Frittu is live on iOS. The foundation is the fridge scan, the personalised recipe generation, the meal planner, and the shopping list. The things I care most about getting right in the next phase are the personalisation engine (getting smarter about what individual users actually like based on what they save and cook) and building out the post-launch content ecosystem around the app.

There's also a Meal Scanner feature in the backlog that I'm genuinely excited about — it goes beyond the fridge and into the full meal identification space. That comes after the core product has found its audience.

If you're a solo developer thinking about building something: the hardest part isn't the code or the design. It's the sustained conviction during the long middle stretch when nothing external is validating what you're doing and you're going entirely on your own belief that the problem is real. That stretch is where most projects die, and it's the part no one talks about honestly enough.

The problem is real. Frittu is the answer I built for it.

Download Frittu on the App Store

Frittu is a free iOS app built in Melbourne. Scan your fridge, get personalised recipes, plan your week, and generate your shopping list automatically. frittu.app