Smart Align

The photography app that teaches composition

while you're adjusting the frame.

The photography app that teaches composition

while you're adjusting the frame.

Designed solo. Built to iOS native standards. Validated with 30 real users.

Final Designs

Read the story ->

TLDR

Short on time? Here's the shot.

Context

Powerful apps exist, none of them teach while you're shooting.

Mobile Photography Gap

Problem

Poor framing, no in-context guidance. Guidance always arrives after the image is captured.

Solution

Real-time AI composition signals that live inside the viewfinder, not a separate screen.

Research

Users

Tested across two photographer profiles (Beginners and Professionals), to understand their approach towards analysing the frame and settings.

Results

Usability Tests showed 6 out of 10 user oriented frames faster within 1 minute and said they can see how the values impact the results.

See Final Designs

Context

Powerful apps exist, none of them teach while you're shooting.

Mobile Photography Gap

Problem

Poor framing, no in-context guidance. Guidance always arrives after the image is captured.

Solution

Real-time AI composition signals that live inside the viewfinder, not a separate screen.

Research

Users

Tested across two photographer profiles (Beginners and Professionals), to understand their approach towards analysing the frame and settings.

Results

Usability Tests showed 6 out of 10 user oriented frames faster within 1 minute and said they can see how the values impact the results.

See Final Designs

I've missed shots. Not because of light or timing, because I didn't know where to look. Turns out, most people haven't either.

Mobile Photography Is Everywhere and they're doing great

but did they address the problems?

Who I Designed For

Two Levels of Photographers.

One Frustration

Beginner

Priya, 22

College student · shoots daily on iPhone

Shoots every day on iPhone. Knows something's off in the frame can't say what. Doesn't want a course. Wants the camera to show her in the moment.

Her frustration directly shaped the alignment signal bar, a single ambient cue that requires zero prior knowledge to read.

Professional

Arjun, 34

Software engineer · Travel photographer

Knows rule of thirds, nothing beyond it. Switches between three apps to get one shot right. Wants the guidance built in, not scattered across settings menus.

His frustration directly shaped multi-grid mode compositional frameworks available without leaving the viewfinder.

Affinity Mapping

Five pain clusters. One root cause.

Five pain clusters.
One root cause.

Every frustration pointed to the same moment guidance that arrives after the shot is already gone.
Not a hardware problem. A timing problem.

Lack of Control

Can't adjust ISO, exposure or aperture quickly miss the moment searching settings.

Composition & Framing

Struggles with straight horizons, subject placement and grids beyond basic 3×3.

Image Quality

Blurriness, poor zoom, hardware constraints leave users dissatisfied
with results.

Dissatisfied with Auto

Automatic filters produce results that don't match what the user imagined.

Contextual Inquiry

Time pressure, multi-app friction and forgetting to check the shot, repeated failures.

How Might We

Teach composition without making the camera feel like a classroom?

The apps already exist. Here's where they stop.

Snapseed

Lightroom

Spectre

Halide

Overwhelming Interfaces

Overwhelming Interfaces Many camera apps (Spectre, Halide) offer advanced manual controls (ISO, shutter speed, etc.), but they overwhelm casual users. Beginners abandon these apps quickly.

Hidden or Scattered Features

Native apps like Apple/Google Camera hide useful features (grids, exposure adjustments) deep in settings, making them inaccessible for quick, everyday use.

AI Exists, but for Glamour, Not Guidance

Apps like Spectre use AI for effects (background blur, long exposure), but none use AI for educating about framing, perspective, or composition rules.

Seven sketches across three sessions.

The goal wasn't refinement, it was getting the wrong ideas out early so screen time wasn't wasted on them.

Finalised screens based on the users feedbacks

Screen 1:

Activity & Progress The home screen surfaces challenge completion and photo count, not for vanity, but because testing showed users needed a reason to return.

Motivation through visible progress, not push notifications.

Screen 2:

The Camera, where everything happens Live exposure data at the top. Multi-grid mode overlaid directly on the viewfinder. Signal bar that turns yellow on alignment. No settings menu. No secondary screen. Every interaction happens without leaving the shot.

Dark UI follows HIG media capture conventions, not a stylistic choice, a platform decision. High-contrast overlays maintain 4.5:1 ratio against the live camera feed across lighting conditions.

Screen 3:

Community Screen, that contains the resources to learn about the basics of the mobile photography, mainly the dynamic challenges with an explanation.

This screen also contains the work of fellow photographers who has posted to the community, help users to get the motivation.

Product Features

Smart Align combines cutting-edge AI with intuitive design to revolutionize mobile photography composition

Image Data in Real Time

While you point out the camera to a subject it analyses and displays the values in real-time and lets users know how the values change dynamically based on light while you pan the camera.

Real-Time Guidance that signals user to align the frame

While user switch into grid-guidance mode they can witness a signal bar, signaling user whether the frame has been aligned as per the selected grid-format. Once the frame is aligned the gray bar turns into yellow, signaling the user that frame is positioned perefect!

What if users can control the live data?

As mentioned earlier, the live data will be displayed at the top of the screen while the camera pans around to let users know what's happening with the image while capturing the image.
But, what if users can control the live data to set specific values while capturing so that can achieve desired results. And this feature delivers exactly the same.

Design Systems

My Goal is to keep UI minimalistic with user friendly layout and not to compete with the the shot.

What I'd do differently?

Designing for when the AI gets it wrong

Low light, glass surfaces, awkward angles. I want to explore how the guidance behaves under real conditions, not just clean ones. That failure state deserves its own interaction pattern.

Taking it beyond Figma.

The interactions are prototyped. I'm looking forward to making them functional, seeing how the design holds up when it's actually running.

Building it with right tech stack.

Swift, SwiftUI, AVFoundation. I've scoped what it takes. Rapid prototyping with the actual tech stack is the logical next step — that's where the real learning happens.