2026 / UC Berkeley Human-AI Design
Stride Recover
Human-AI design portfolio for a smart compression sleeve that supports hamstring recovery and return-to-play confidence
Overview
Stride Recover is a low-friction smart compression sleeve concept for recreational athletes managing hamstring strain recovery without regular clinical supervision. The project combines human-led interviews and observations with AI-augmented synthesis, semantic mapping, concept generation, and prototyping to convert muscle monitoring into clearer recovery guidance.
Problem / Context
Recreational athletes often self-manage hamstring injuries with inconsistent routines, low access to physical therapy, and uncertainty around safe return-to-play timing. Existing solutions split between simple tools with little feedback and clinical systems that are too expensive or disruptive for everyday users.
Role
Team project - user research, AI-augmented synthesis, concept selection, wearable interaction framing, prototype evidence, and portfolio build
Institution
UC Berkeley Human-AI Design
Team
Loris Emanuelli, Human-AI design teammates
Tags
Human-AI / Wearables / Sports Engineering / User Research / Prototyping
Process
- - Conducted semi-structured interviews and warm-up observations with recreational athletes, gym-goers, and intramural players
- - Mapped social, economic, and technological factors around injury management, confidence, access, and wearable sensing
- - Used AI for semantic mapping, competitive framing, and concept divergence while keeping problem selection and judgment human-led
- - Compared concepts against setup effort, comfort, real-time feedback, prototyping feasibility, and return-to-play support
- - Converged on a smart compression sleeve using EMG-style muscle monitoring and simple actionable feedback
Key design decisions
- - Prioritize a familiar sleeve form factor over a clinical-looking device to reduce adoption friction
- - Translate raw muscle data into confidence-building guidance rather than exposing users to ambiguous sensor readings
- - Frame feedback around bilateral activation and recovery readiness so athletes can understand asymmetry and risk
- - Use the class portfolio as a transparent record of where AI helped and where human research drove the project
Engineering details
- - Prototype direction combines compression garment ergonomics, wearable sensing zones, feedback logic, and mobile/dashboard storytelling
- - Design evidence includes interview synthesis, product opportunity gap selection, poster artifacts, final report visuals, and prototype demo media
- - Human-AI workflow separates human-only research, AI-augmented synthesis, AI-generated ideation, and final human judgment
- - Concept is designed for daily activity context rather than clinic-only measurement
Outcomes
- - Delivered the final Human-AI project site and a complete visual design portfolio
- - Built a coherent concept story around accessible recovery support for recreational athletes
- - Produced poster, report, logo, video, and prototype artifacts for course submission
- - Turned the class project into a reusable case study for sports wearables and human-AI design process
Gallery
Video
What I would do next
- - Build a higher-fidelity sleeve prototype with validated sensor placement and washable electronics strategy
- - Test feedback wording with athletes recovering from recent hamstring strain
- - Define recovery-readiness thresholds with physical therapy or sports medicine input
- - Run a short longitudinal study comparing adherence and confidence with and without feedback