This morning I decided to mess around in Google AI Studio — partly out of curiosity, partly because I wanted to witness the power of Gemma 4 for myself.
This is also my first proper attempt at coding in months, since that disastrous HNG internship experience. For context: HNG is a competitive tech internship where I was up against roughly 5,000 other Africans all vying for top spots. Despite paying for premium access, I dropped out after Stage 3. I felt out of place and overwhelmed — and honestly, that knocked my confidence for a while.
Finding My Footing
On the professional front, I'm still figuring things out. As someone who's neurodivergent, navigating the tech world comes with its own particular set of challenges.
I graduated with a Computer Science degree in 2017. And yes — before anyone brings it up — I know the "degrees don't matter in tech" debate well. But there's a cultural layer here that matters: in Nigeria, a degree still carries significant weight. It signals credibility and worth, especially to the older generation.
There's also the NYSC — the mandatory National Youth Service Corps that Nigerian graduates are expected to complete. Due to complications with my online university programme, I wasn't able to participate. The officials told me there were simply no provisions for my situation. That particular chapter is a story for another day.
Back to Building
A few months ago, I used Mocha AI to generate a skincare tracker app. I wanted to revisit that concept — this time using Google AI Studio — and bring my product management skills into the process. (I got a product management certification through Treford three years ago, and it turned out to be quite useful here.)
My goal: develop a personalised skincare routine tracker that helps users manage, optimise, and maintain their skincare habits. The app lets users log products, track routines, set reminders, and monitor skin progress over time. AI-driven insights offer suggestions based on skin type, concerns, product ingredients, and goals. The system also accommodates users with varying skincare knowledge through visual guides, tutorials, and accessible explanations.
Tech Stack: Python · JavaScript · React Native
Key Features
Milestone 1: Personalised Skincare Routine Engine
AI-powered routine recommendations based on skin type, concerns, age, and goals
Product suggestions based on ingredients and compatibility
Routine adherence tracking with progress insights
Product logging (cleanser, serum, moisturiser, etc.) with frequency recording
Progress photo uploads to track skin changes over time
Daily and weekly reminders for routine steps
Milestone 2: Contextual Customisation
Visual tutorials for product application (e.g. how to layer a serum or apply SPF correctly)
Beginner- and advanced-friendly explanations
Improvement highlights, missed-step flags, and trend-based tips
Conflict alerts — for example, flagging the combination of retinol and vitamin C used simultaneously
Milestone 3 (Advanced / Optional)
AI analysis of uploaded skin photos to detect changes like acne, dryness, or pigmentation
Ingredient checker for potential irritants
Personalised product recommendations for future purchases
Who Is This For?
The app targets skincare enthusiasts, beginners, and anyone managing specific skin concerns — particularly users who want to build consistent routines without being overwhelmed by information overload.
It's still a work in progress — I'm actively tweaking features and improving the experience. But getting back to building something, even something small, feels like the right step forward.
Top comments (0)