Process starts here ↓
The TLDR ↓
I explored various formats for the main screens. Although there were preset layouts in the application, I tried to keep my mind open to new ideas at this stage.
I made the 'Camera' page the main interface, basing all design decisions on research while following the design system.
The high-fidelity Camera page focused on fitting in closely with the design system, and keeping all essential information visible or available in one click.
Starting off by merging into the existing designs.
Crafting the ideal camera information page.
Design • Ideation • Interaction Design
Iteration 1
Iteration 2
Iteration 3
Final Mid-Fi of Camera Page
Camera Page
View live feed
Edit Camera details
Edit alert sensitivity
Edit alert objects
Shortest Flow
Landing page
Click card
Preview of
leopard detection
Allows user to double check information when taking action.
1
Just one click from Landing Page to Previewing the footage of a possible leopard attack. The Leopard Detected card only appears when there is a possible leopard attack.
Through Camera Flow
Landing page
Specific camera page
Learn more
Watch live
Watch feed of camera
1
2
Can learn more about the camera that detected a possible leopard attack directly from the Landing Page card. Important to find the location of the camera if that is not clear.
Through Devices Flow
Landing page
Security cameras preview
Devices Page
Devices
Security Cameras
1
2
4
Choose camera
Specific camera page
Watch live
Watch feed of camera
3
The primary flow is through the Devices tab that exists in the parent app. This is necessary because the primary focus of the app is not this feature.
The key interaction - viewing the
Live Feed - was accessible through multiple entry points.
The new feature's key aspects formed the foundation of a sub-design system.
Too little space for all details
Must go back to previous page to switch cameras
Unnecessary step
Way to edit Details/Alerts is not clear
Don’t rely on the user clicking around
Can switch between cameras without going back to previous page
Can identify camera source without reading
Reduces cognitive load
Improved information hierarchy
“Watch live” is familiar language
Reduces confusion about “feeding”
Card-style interaction is unnecessary
Main use of page is not prominent enough
Confusing term
“Feed” could mean multiple things
Prioritizes Details over Live Feed
Information hierarchy should be improved
Cannot identify camera source without checking feed
Unnecessary step
Clear Edit button instead of card flipping
Clearer calls to action
Nearly all interactions were a simple Slide In / Slide Up to reduce the amount of motion on the screen.
Creating interactions that are short, but seamless.
VALUE PROPOSITION
UX FEATURES OBSERVED
Ability to set up and view feed from multiple cameras, Indian users are target audience
Familiar formats with video on top and information below
Color markers for important events and replay
Horizontal scroll to see past events
VALUE PROPOSITION
UX FEATURES OBSERVED
Security camera with integrated support, Indian origin, personal and industrial use case
Tabs on bottom or top of screen
Preview of camera helps see location without reading name + location of camera
Can message/call for help
VALUE PROPOSITION
UX FEATURES OBSERVED
Security camera + doorbell with integrated UI, United States origin, for personal use
Device feed and device details on single page
Customizable alerts, customizable sensitivity to images + sounds
Icons without captions are confusing
The closest existing solutions are smart doorbells and surveillance systems.
Our user highlighted the apprehension of farmers, and the need for 24/7 support.
Research • Design Tenets
Indian brands use standard layouts, because they have the fastest adoption for Indian users.
We also pitched making the MVP product & related support free for farmers.
Knowing about leopard attack too late
Worry about support for tech issues
Fear of complications with set-up
Worry about costs + maintenance
Unsure about accuracy of technology
PAIN POINTS
Fast system:
minimal clicks from alert to action
Make alert system the focus of the application only when needed
Use existing design system
Highlight and provide multiple channels of support
Allow for verification of event
Allow user to adjust aspects of alert system
DESIGN TENETS (New)
FROM PARENT APP
Opportunity
24h surveillance?
Requires staff to observe cameras that farmers cannot afford.
TOO EXPENSIVE
Livestock enclosures?
Severely limits livestock to enclosed area - quality of life affected.
UNETHICAL
Chemical leopard repellant?
Can cause leopard death.
UNETHICAL
TECH FOR GOOD
Smart alert system?
Informs farmer when attack is imminent.
OUR SOLUTION
Indian farmers are unable to protect themselves against leopard attacks.
Why is this happening?
Unfortunately, most solutions cannot solve this problem without dire side effects.
Leopard attacks on farms are rising due to habitat loss from deforestation, forcing them to seek prey on nearby cattle farms.
Our research team found that farmers were unable to respond quickly, losing cattle and other farm animals often.
Imagine you're just chilling on your farm when a leopard attacks your cattle. You rush inside to save yourself, but your cattle are left to die.
Sounds like a wild story, but this is a real problem affecting farms in India.
Problem
What we achieved ↓
This is the end of the case study!
Thank you for reading!
Always up for a coffee (no sweetener).
Let's connect!
A fast alert system with constant support.
Our system benefits farmers by allowing them to take proactive action.
Farmers are alerted of when an attack is about to happen, giving them enough time to move their staff and animals inside, saving lives.
Solution
Preview of captured feed.
Allows farmers to accurately locate leopard sightings and take informed decisions.
Customizable alerts.
Offering farmers flexibility filters out non-essential information, reducing alert fatigue.
Support anytime.
24/7 support through calling/text/chatbot integrated throughout application.
Outcomes
Reflections • Future Steps
Ability to move quickly from alert to action.
Support across platforms and throughout the application.
Customizable alerts and notifications.
After the internship, I was dejected that usability testing was not completed. However, this happens often in industry projects with low-budgets. I learned to adapt to the process that was feasible.
Most farmers interface with English-based applications like Youtube, Whatsapp, and Facebook. However, to continue diversifying the app, we need to add more languages to the interface.
But, there are so many languages in India! It won't be easy.
Yes, the farmers took time to be comfortable with the product. But, they still had high expectations for the quality of design. They showed clear preference towards clean UX in the research, and we made sure to try to fulfill that in our design.
A fast, intuitive, and supportive feature.
Testing on 100+ farms in India today.
Even the simplest flow can take
time!
Don't assume that people who use technology less frequently lack technical skills.
In the future, we have to conduct extensive testing and incorporate many languages into the interface.
So, let's get on to the technical feasibility.
One major issue was that we didn't have many images of Indian leopards to begin with.
Implementation • System Design • AI Model
With limited data, the AI model wouldn’t be highly accurate at first. However, we designed a system that ensures the product’s longevity, allowing it to improve over time as more data becomes available.
When the phone app processes videos of leopards, they are sent back to the database to inform the AI model in the future.
Model
User-facing
IOT Device
Jio Platforms required all our MVP elements to be ready, so I used my programming background to create the AI model to detect dangerous animals - focusing on leopards.
Initial System Design
Final System Design
IOT DEVICE
CAMERA
SOLAR PANEL
DATABASE
AI MODEL
PHONE APP
AI MODEL
IOT DEVICE
PHONE APP
100+
FARMS IMPACTED
AI Alert System for Safety
An end-to-end solution using AI to power a real-time alarm system, ensuring the safety of farmers and their livelihoods.
Context
Product Design Internship
Jun - Nov 2023
Team
Individual + Product Manager
Responsibilities
Competitive Analysis
Journey Mapping
Data Cleaning/Augmentation
AI Engineering
Sketching
Prototyping
Tools
Figma
FigJam
TensorFlow
Python
Google Colab