This project did not start with AI.
It started with a very simple and frustrating question I kept asking myself almost every day. I have clothes, so why does deciding what to wear still feel like a task?
Most days the problem was not about fashion. It was about context. Weather did not match the outfit. The clothes did not feel right for where I was going. Sometimes everything technically worked, but the combination just felt off. And mornings are not the best time to experiment.
When I started thinking about this from a product and technology perspective, one thing became clear. Outfit decisions are not random. They are contextual.
You already have all the inputs.
Your wardrobe.
The weather.
The occasion.
Your personal preferences.
But there is no system that brings all of this together.
Why this is a technology problem
At a high level, outfit selection is a decision making problem.
You are choosing a combination of items under constraints. Weather constraints, occasion constraints, personal comfort, and visual compatibility. Humans do this intuitively, but we are inconsistent and biased. We fall back to safe choices and repeat outfits because the cognitive load is high.
This is where technology actually makes sense.
Not as a fashion trend engine or a shopping recommender, but as a decision assistant.
The core idea behind DripComb is simple.
Use AI to reduce decision fatigue, not to replace personal style.
How the system thinks
From a technical point of view, the problem can be broken down into three parts.
First is understanding what exists in the wardrobe. Clothing items need to be categorized and tagged in a structured way. Things like type, color, season suitability, and formality are more useful than brand names.
Second is context. Weather data, occasion selection, and basic user preferences act as constraints. An outfit that works in winter should not be suggested on a hot day. Something casual should not be pushed for a formal setting.
Third is ranking combinations. Not every valid combination is a good one. The system needs to score outfits based on compatibility and past user feedback. This is where learning happens over time.
The goal is not to generate hundreds of outfits. It is to suggest a few that actually make sense for that specific moment.
Why this is not about shopping
Most fashion related apps push discovery and buying. That works for commerce, but it does not solve the daily problem.
DripComb is intentionally focused on using what the user already owns. Better combinations lead to better usage of existing clothes, which is both practical and sustainable.
In many ways, this is closer to a personal productivity tool than a fashion app.
Current stage
Right now, DripComb is in the early MVP stage. The focus is on getting the core logic right rather than polishing everything. Understanding how people respond to suggestions, what they reject, and why they reject them is more important than visual perfection at this point.
This is being built in public, with iteration driven by real feedback rather than assumptions.
Why I am sharing this early
Writing about this helps clarify the problem and the solution. It also helps connect with people who have faced the same issue, whether as users, designers, or engineers.
If you have ever stood in front of your cupboard feeling confused despite having clothes, you already understand the problem this is trying to solve.
If you are curious about the product or want to follow along as it gets built, you can join the waitlist at dripcomb.in. Early feedback genuinely helps shape what this becomes.
This is still early, but the problem is very real._
Top comments (0)