My dreams of becoming a pro-level developer have unsuccessfully never gone beyond basic HTML.
I can structure a page and close my tags properly. But anything concerning Python and full-stack frameworks has proven impossible for my neurodivergence.
Google’s experimental AI tool has revived my hope. Opal removes most of the technical friction involved in coding and lets me bring apps to life with mere descriptions.
I’ve built two interesting apps with it on my personal computer, and they work. Here’s how.
12 AI prompts I use to get more done in less time
I used to spend hours on my tasks until I tried these prompts
Cook with whatever is in your pantry
Meal prep works with your budget, not against it
I had the idea of building an app that helps people cook with what they already have. Most meals are planned, and you’ll follow a recipe to prepare them.
But there are moments when you’re short of one ingredient or have limited time. It may cross your mind to improvise.
I wanted my app to make the improvisation cost-effective and prevent wastage. The official Opal platform provided the right workflow to make it a reality.
It provides color-coded tabs that represent user input (light green), generated prompts (neutral grey), output (green), and adding assets.
The User Input tab created the fields for my app. It added input blocks for Ingredients List, Dietary Restrictions, Energy Level, and Cooking Time.
The Generate block gathered and processed all the inputs. I instructed the system to prioritize ingredients that are about to expire, respect dietary restrictions, and adjust complexity based on energy level.
The Output block then determined how the response appeared. Its task was to render the meal name, prep time, simple steps, and an explanation of how users would minimize waste.
Tweaking these tabs isn’t necessary. If the output isn’t what you want, 80% of the time the fix is in your prompt. You only need modification if you remember that your app also needs additional fields.
My example prompt: You are a smart food optimization assistant. When a user provides a list of ingredients, especially items close to expiring or with dietary restrictions, you must prioritize them. Suggest one to three realistic meals and minimize additional required ingredients. Keep instructions clear, and don’t suggest complicated recipes or specialty ingredients unless absolutely necessary.
It took roughly a minute for my web app to generate, after which I was able to preview it.
In my mock test, I listed expiring ingredients, such as spinach with two days left, bell peppers, cooked rice, and Greek yogurt.
My available staples were eggs, garlic, and olive oil, with my dietary focus being on the exclusion of pork and using low dairy. The meal was to be prepared in under 20 minutes.
My software generated a recipe for a speedy spinach and bell pepper egg scramble with rice. It also provided explanations for why it fit into my plan.
Find the right shade for you
Beauty product matches go skin deep
My next app was a personal project. I’ve always struggled with finding the right foundation shade for melanated skin. The undertone variations are wider, such that they may lean toward golden, red, olive, neutral, or even slightly cool.
Many beauty brands compress all of them into one or two warm categories, which isn’t accurate enough. My app idea is called Melanin Match, and it was inspired by one of my favorite apps on the Play Store, Unotone.
But mine is a more toned-down version, and still needs refinement. You’ll upload a photo, and the AI scans your face. Then it generates curated foundation, concealer, blush, and lip suggestions.
My results revealed I have a deep, dark, and warm skin tone. A lot of the recommendations it generated are shades I haven’t bought yet. But I have run multiple image searches, and looked at comparison photos. The proposals looked like close matches.
However, some of them, like the Fenty Pro Filt’r Soft Matte 450 foundation, are expensive. It’s forced me to refine the app and add a budget filter. That way, the AI wouldn’t default to premium brands only.
6 things I had no idea Gemini could do
Google Gemini just got even more useful for me
Give your creativity a trial run
Opal apps aren’t traditional mobile apps you can package and upload to the Google Play Store. Instead, when your app is ready, you can publish and share it via a link. Anyone can access and use it on a browser.
The platform is an amazing way to test ideas quickly and see how they perform in the real world. You won’t have to invest months in frontend and backend engineering before validating whether your concept is useful.
You’ll also identify where the experience feels subpar. When you have the time, it’s also worth exploring apps inside Google’s experimental ecosystem. Beyond Opal, the Mixboard cracked my style, and has become another favorite tool.
