Found&

Overview

Found& is a desktop app that helps creatives save inspirations and turn them into a creative direction, capturing not just what they find, but why it matters and where it could lead.

I was the sole product designer to define and prototype the MVP for Found&. Our primary objective was to find out whether the way people saved, annotated, and collaborated on inspirations could really help people turn what they collect into something they act on.

---

The problem

Creative inspiration is rarely just a pretty picture. But most inspiration tools limit people to passively collecting images and videos from a feed, rather than understanding, developing, or acting on what they’ve saved.

Looking at how users currently save and organize their references, we identified three major problems in existing tools:

Organizing multimedia references

Inspiration can come from anywhere. A single project can use a blend of visual references, documentation, videos, articles, music, code, and many more. However, capturing diverse media across scattered platforms is difficult to organize and retrieve centrally.

Loss of context

Without context, saved inspiration loses its thread the moment the initial insight fades, making it hard to develop ideas and turn them into direction. That same gap means collaborators have to reconstruct intent before they can build on each other’s finds, instead of picking up right where a save left off.

Algorithmic discovery

Existing platforms rely on engagement-driven, visual-similarity algorithms that dictate our inspiration for us. By forcing us onto visually similar, homogeneous content, these algorithms trap us in creative echo chambers.

---

Opportunity

While existing tools addressed isolated fragments of the workflow, there were little to none which offered a holistic solution.

The gap wasn’t collection, it was everything after it, the in-between space where a save either becomes direction or dies in a folder. We wanted a system that could hold any kind of inspiration and carry enough context to survive that trip, moving the way real references do — passed along and reworked, shaped by taste rather than an algorithm serving up more of what you’d already seen.

---

Defining scope

To quickly validate our core concept, we had to scale back on our massive vision of a limitless multimedia ecosystem featuring an infinite-canvas moodboard into a highly focused MVP.

Desktop first

Inspirations and resources are often spotted on mobile, but utilized on desktop. We prioritized launching a desktop application paired with a browser extension, ensuring that saved resources and their context remain accessible directly within the user’s workflow. A browser extension was chosen as it was the most direct method to save what users come across while browsing online.

Core features

We prioritized essential features for capture (web clipping, intention inputs, folders, source retrieval) and planning what to do with saved inspirations (following feeds, randomized discovery, shared folders).

Supported media

We restricted V1 to common media formats such as images, videos, GIFs, links, text, and full-page snapshots to ensure a lean, testable build without sacrificing the multimedia experience.

---

Solution design

Here is how I designed the most essential features:

01. Cards

Card design. The main challenge was displaying diverse media types of varying sizes and orientations. Instead of displaying them in their respective orientation and sizes, I unified all assets into square thumbnails with the user’s intent displayed underneath. This required a trade-off of not being able to see the uncropped asset at a glance for structural consistency. This established a predictable visual rhythm where users always know exactly where to find the context they are looking for in a sea of varying media.

Collections. The cards were organized into a folder system called “Collections”. I used a spacious grid format, giving each inspiration enough room to be isolated during quick scanning. I also introduced a “Quick Save” default collection to capture on-the-go inspiration without forcing immediate categorization.

Expanded view. To solve the trade-off of the cropped thumbnails, I designed an expanded viewer. This viewer displays the asset in its original, uncropped form alongside crucial metadata. Most importantly, it features the source link, making it easy for users to retrace an artifact back to its origin.

02. Feeds

Following feed. This feed was designed to surface artifacts saved by users within an individual’s network, allowing them to discover and adopt artifacts from trusted users. A major challenge was preventing feed fatigue. If someone saved twenty images in a row, it would overwhelm their followers’ feeds with twenty posts. To solve this, we designed a system that groups artifacts into a single post based on what collection they were saved to, and the timeframe of the saves.

Post design. Posts were stripped back to the absolute essentials: who saved the artifact, and what folder it lives in (if it’s public). The only actionable items for a user are hover to save to their own collection or click to see the expanded view of the artifact. I intentionally chose not to implement likes or the number of saves because the focus was to foster a space of utility and creative workflow, not create an environment of social validation or performative curation through metrics.

Discover feed. For organic discovery that is not guided by algorithms, I designed an algorithm-free feed that completely randomizes saved artifacts from users outside of an individual’s immediate network. I utilized the same grid of cards collections use, for quick thumbnail scanning and deep dives via the expanded view. I also added a refresh button to give users agency over when they can get new content, rather than automatically refreshing the page every time they enter it.

03. Media capture

Context-aware capture methods. To handle various formats of media seamlessly, I designed a browser extension with multiple entry points.

Visual preview. Since webpages can have numerous elements, I added a visual preview of the saved element at the top of the capture modal to eliminate any uncertainty about what asset is actually being grabbed.

Input fields. I simplified the input fields to the essentials: collection selection, save intention, and sharing options. While it would align with our goal of mindful curation to make collection selection and writing save intentions compulsory, this could introduce significant frustration and friction. I compromised by making the intention field optional and implementing a smart-default to a “Quick Save” collection.