Sunny Ray talks with founder Vladimir Dubrovsky about the daily problem of deciding what to cook. After losing his job, Vladimir took over the family kitchen and found that the real pressure was the decision, not the cooking. He has almost three decades in high tech as a project manager, product manager, and QA leader, but had never shipped production code. He nearly dropped his app idea when a former engineering colleague could not commit. He then asked ChatGPT whether he could build it alone, received a step-by-step roadmap, and had a working prototype within months. The conversation covers his shift from copying code to directing AI like a director, and how tester feedback reshaped recipe quality and led to US and metric ingredient lists. It also covers how Meal Tap suggests dinners by mood, plans the week, builds grocery lists, and guides cooking. Vladimir explains how per-meal servings help households with different tastes, describes an experimental restaurant photo to home recipe idea, and says his next 90 days will focus on marketing and distribution. Sunny pitches a fridge photo feature and a home kitchen marketplace.
A laid-off high-tech veteran with no production coding experience used AI to build an app that ends the nightly question of what to cook.
What problem are you solving, and where did the idea come from?
I found myself out of a job and started cooking for my family. After five meals in a row I realized the hard part was the daily decision and the pressure that comes with it. Before, that was all my wife. Once I was running the kitchen, I started wondering how to decide what to cook every day.
Why did the idea almost stop when you were building with a former colleague?
I'm not a developer. My career was project manager, product manager, and QA. I needed someone to build with, but my former colleague couldn't commit, so I almost gave up. Then I asked ChatGPT whether I could do this alone. It said yes, given my experience, and gave me a step-by-step roadmap that made sense.
How did your relationship with AI change as you built the product?
At first I was just a programmer, copying code and fixing compile issues. Then the roles kept switching: product manager, project manager, architect, marketing specialist. My work moved from writing code to making decisions and discussing architecture. Now I feel like I'm sitting in a director's chair, managing AI agents and having them do what I want. It's challenging, but fascinating.
What user feedback changed how you think about the cooking experience?
Early testers gave me lots of positive feedback, but the most valuable was about recipe quality. One person showed me that ingredients must be written in a specific order, so I redesigned my recipe pipelines to produce professionally structured recipes. I also added ingredient lists for US or metric markets, because measurements that work in one region don't work in another.
How does Meal Tap work, and who is it for?
Meal Tap helps you make a decision. You tell it how you feel, like cozy, fresh, or quick, and it suggests dinner. Then you can plan your week, build a grocery list, and cook step by step. It's free to start. I want it to become a personal cooking assistant that knows your preferences and what's in your kitchen.
How does Meal Tap handle households where everyone wants something different?
I believe I solved that. When you plan a meal, you choose how many servings it is, so you can plan another meal alongside it. For example, one person has fish and two people have steak. That gives everyone a dinner they want. It's still something I keep thinking about, because family preferences really do differ.
What have you learned about finding users, and what are you focusing on next?
Getting attention on social media is easy, but getting people to install the app and use it daily is hard. I spent a lot of time building the product and not enough on acquiring users early on. Over the next 90 days, maybe less, I'm shifting to marketing and distribution, including finding creators and a clear message.
What's the hardest part of turning a restaurant photo into a recipe you can cook at home?
The hardest part is making sure the AI can read the image and fit the result into the Meal Tap recipe structure. I travel a lot and photograph dishes at restaurants, then wonder if I can cook them at home. I've run experiments in a playground I built for this, and I think it could be realized soon.
Building something daring? Sunny talks to founders like this every day. Fifteen minutes to see if your story belongs on the stage.
Claim your pre-interviewSend a founder here and we will know it was you.