Soulside AI - Cohort 19
Soulside is a HIPAA-compliant mental health platform that combines telehealth capabilities with AI-powered documentation to streamline therapy sessions and improve patient care.
As part of Cohort 19, our team is focused on designing the Voice Enrollment and Recognition System for Soulside. This feature enables automatic identification and separation of speakers during individual and group therapy sessions, powering accurate AI-generated notes and transcripts. Alongside technical implementation, we’re also refining the system’s usability by establishing a clearer, user-centered design language and adapting the PRD in response to real-world friction points.
We made transcript wait-time more enjoyable with micro-animation
The side-by-side glow up
Soulside glow up 101
Why we're excited
As designers ...
We designed at the intersection of AI and mental healthcare, where design conventions are still emerging. This allowed us to shape foundational patterns and introduce a more human-centered design language. Our challenge was making advanced AI feel intuitive, transparent, and trustworthy—bridging the gap between complex technology and real user needs.
As product visionaries ...
Soulside is transforming mental healthcare by ethically integrating AI into human-led therapy. The product address critical pain points from documentation overload to access disparities. It doesn't just streamline clinical workflows—it amplifies empathy and insight in a field that needs scalable, human-centered solutions.
The how
01
02
03

THE BEGINNING | Our first 3 weeks
Since many of us were new to healthcare and IOPs, we took several steps to build our understanding. We started with domain research, familiarizing ourselves with telehealth, clinical workflows, and important terms like SOAP notes and HIPAA compliance. To better understand the needs of therapists and providers, we conducted interviews to learn about their current workflows and how Soulside AI could integrate into their routines.Instead of just comparing similar products, we audited the best tools across telehealth, transcription, and AI scribes to identify what works well and what doesn’t. We also tested these tools ourselves during meetings, which helped us spot pain points and opportunities for improvement. Finally, we conducted a usability audit of the existing platform by role-playing therapy scenarios. This gave us valuable insights into the product’s strengths, gaps, and potential to support clinicians effectively.

ideate, constrain, & test
We assessed the usability gaps and identified areas for improvement in the existing user flow and sitemap from the internal usability audit. We used Crazy 8’s and a co-ideation session with our client to generate and refine ideas for the MVP. Next, we pinpointed three critical user journeys to prioritize and began prototyping around them. We also explored design principles focused on trust—both between users and AI, as well as between users themselves—to ensure Soulside could foster confidence with clinicians. A mid-sprint review allowed us to align closely with the client, narrow our scope and define an MVP (minimum viable product), and streamline our design direction for a more cohesive and consistent approach moving forward.

innovate, spec, & ship
In the final stretch, we merged our three user journeys into one core Critical User Journey: helping Soulside clinicians access and edit notes and transcripts with full transparency and trust in Soulside’s AI. We focused on high-fidelity design iterations and ran usability tests with clinicians and designers to uncover what worked and where to improve. Our team emphasized what made the experience stand out—intentional design language, thoughtful animation, and voice transcription features that bridge UX and AI engineering. We also workshopped our storytelling skills and began turning key design assets into a polished slide deck that reflects both our design thinking and the system we created for Soulside.
Impact & shipped
Build efficiency into clinician’s workflow with bird-eye appointments management
What is the future of AI native interface and how do we bridge the gap?
• Designing workspaces that morph based on context vs static documents
• Canvas adapts in real-time to your workflow - not just another doc editor
• Content self-organizes based on relationships vs manual folder structures
• Information flows between blocks naturally, like liquid vs rigid containers
• Elements reshape and connect based on AI understanding of your work
• Built for flow state - content appears where needed, then vanishes
• Designing workspaces that morph based on context vs static documents
• Canvas adapts in real-time to your workflow - not just another doc editor
• Content self-organizes based on relationships vs manual folder structures
• Information flows between blocks naturally, like liquid vs rigid containers
• Elements reshape and connect based on AI understanding of your work
• Built for flow state - content appears where needed, then vanishes

Cater to the clinician’s convenience when running IRL group sessions - no-click-needed UI
Make patient data pretty with a clear, soul-centered visual language
Build trust with interface that clearly communicates constraints
Some of our team memories

First team kickoff — Week 1, where it all began!

The Great Gradient Obsession – the UI battle we never meant to start but couldn't stop once it began

The only cohort team whose all members show up IRL at CoCreate Conference 2025! And once again we rock the storytelling ^^