Project Info

Deskercise

Devpost

Inspiration

We often find ourselves hunched over our laptops and looking at the screen for a little too long, barely moving for hours. Long periods of sitting are proven to cause stiffness, fatigue and poor posture. But realistically, who's going to bust out a full yoga post at their desk at work/school? That's where desk exercises, or Deskercises, come in.

What it does

Users can schedule stretch breaks through their Google calendar or follow a Pomodoro-style workflow. At each break, the app guides them through quick, desk-friendly stretches using real-time pose detection to track accuracy and completion. With voice coaching and feedback powered by LLM, Deskercise makes micro-movement fun and engaging. Users earn points for completing stretches correctly, turning breaks into a gamified wellness routine that fits seamlessly into their day.

How we built it

We built Deskercise using a set of AI-powered and web technologies to deliver a seamless, interactive experience: ReactJS – Built the responsive frontend and UI flow for desktop stretch sessions. Google MediaPipe – Enabled real-time pose detection via webcam to track and validate stretch accuracy. Gemini 2.5 Flash – Used to interpret posture data and generate personalized stretch break recommendations based on user's Google Calendar for the day Vapi – Voice-guided coaching system that delivers real-time instructions and motivational feedback. Google Cloud – Integrated calendar functionality to schedule or sync stretch breaks with the user’s daily workflow. Potentials of Deskercise Deskercise has strong applications in both B2B and B2C markets — bringing smart, desk-friendly movement to a wide range of users who spend long hours seated. 🚀 Use Cases 🏢 Corporate Employee Wellness Initiatives: Offer Deskercise as a workplace wellness tool to promote better posture, energy, and productivity across teams. 🏢 Corporate Employee Wellness Initiatives: Offer Deskercise as a workplace wellness tool to promote better posture, energy, and productivity across teams. ☎️ Call Centers: Help agents combat sedentary fatigue with light, guided stretches during shift breaks — no need to leave their desks. ☎️ Call Centers: Help agents combat sedentary fatigue with light, guided stretches during shift breaks — no need to leave their desks. 🕹️ Gaming Communities : Encourage streamers and gamers to take short, guided breaks to reduce strain and stay alert during long play sessions. 🕹️ Gaming Communities : Encourage streamers and gamers to take short, guided breaks to reduce strain and stay alert during long play sessions. 📚 Study Spaces: Support students during intense study periods with low-effort movement prompts that boost focus and reduce stiffness. 📚 Study Spaces: Support students during intense study periods with low-effort movement prompts that boost focus and reduce stiffness. 💻 Remote Workers / Freelancers: Give home-based professionals the tools to stay active and mindful of their health, right from their workspace. 💻 Remote Workers / Freelancers: Give home-based professionals the tools to stay active and mindful of their health, right from their workspace.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • CSSIn code
  • HTMLIn code
  • JavaScriptIn code
  • ReactIn code
  • Google GeminiClaimed

4 of 5 appear in the indexed code. 1 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

240 KB

Source files

50

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars