Employment · 02
Stealth Startup
Co-founding a personalized skincare platform with working product recommendations, client plans and logs, and iOS and web applications. Near MVP by August 2025, with AI classification still in development.
Explore a schematic mobile prototype and clinic workspace. The detection view uses illustrative scores, not patient data or measured model results. All portfolio information is available in the page text.
Preparing the exhibitMobile routine tracking and a clinic review workspace · illustrative prototype
The work
Personalized skincare, from product data to recommendations
Overview
From May to August 2025, I worked as a co-founder and engineer on a personalized skincare platform. We were building a system that used information about a client’s skin and other client details to recommend products, alongside skincare plans and logs.
The business combined recommendations for existing skincare products with plans for our own line developed through an original design manufacturer (ODM). The recommendation platform, client applications, and product catalog were parts of that broader idea.
My responsibilities
My role covered both building the product and pitching it to investors. I was responsible for frontend development across the iOS mobile application and web application, building the product recommendation platform, and contributing to parts of the AI.
Product data and recommendations
I collected skincare product information to support the recommendation platform and affiliate marketing. The goal was to connect what we knew about a client’s skin and needs with relevant products, including our planned skincare line.
Image classification and facial data
I built image classifiers and worked on the facial-data collection needed to support them, including consideration of privacy and applicable legal requirements.
The training-data challenge
As an unfunded, pre-seed startup, our biggest challenge was building a reliable training dataset with limited resources. We researched image sources across the web, looking for licenses that permitted our intended use and examples supported by clinical assessments.
Finding images was only part of the problem. We also needed reliable labels for classification. A doctor’s assessment of a single photograph could be provisional: additional images of the same person might be needed to establish a more confident diagnosis.
The challenge was finding both usable images and enough clinical evidence to support their labels. That uncertainty made training-data quality a central constraint on our classification work.
Progress by August 2025
We were close to a minimum viable product. The recommendation platform worked, while the AI classification still needed further development.