ShiftSaarthi

• 

  Landing Page

• 

  Dashboard Page

• 

  Handoff Page

Inspiration

ShiftSaarthi was inspired by the communication challenges that happen during patient handoffs in hospitals and clinics. In many healthcare environments, nurses still rely heavily on verbal communication and manual notes while transferring patient information between shifts. During busy hours, important updates can become repetitive, delayed, or unstructured, especially in multilingual environments where communication naturally switches between Hindi, Hinglish, and English.

The idea was to build a practical workflow automation tool that simplifies this process using voice input and AI-powered analysis.

What it does

ShiftSaarthi is a multilingual voice-first healthcare workflow automation platform.

The platform allows healthcare staff to:

-record or type patient handoff updates -convert speech into text in real time -generate AI-powered patient summaries -detect urgency levels and clinical risks -extract actionable follow-up tasks -maintain timestamped handoff history

The goal is to improve communication continuity and reduce workflow friction during shift changes.

How we built it

ShiftSaarthi was built using:

-Next.js -TypeScript -Tailwind CSS -Google Gemini API -Web Speech API

The frontend was designed as a healthcare-style operational dashboard. Real-time speech recognition was implemented using the browser’s Web Speech API, while Google Gemini API was used to analyse transcripts and generate structured workflow outputs such as summaries, alerts, and action items.

Challenges we ran into

One of the biggest challenges was handling multilingual voice input smoothly while maintaining a clean workflow experience. Speech recognition can behave differently depending on accents, mixed-language speech, and browser limitations.

Another challenge was keeping the product simple and realistic without overcomplicating the architecture during the hackathon timeline. The focus was on building a stable and believable workflow MVP rather than adding unnecessary complexity.

Accomplishments that we're proud of

We’re proud of building a working multilingual workflow system that feels operational and practical instead of just another AI chatbot demo.

Some highlights include:

-multilingual speech-to-text support -AI-generated workflow summaries -urgency detection -timestamped handoff history -healthcare-focused dashboard UI -real-time workflow interaction

Most importantly, the platform demonstrates how AI can simplify communication-heavy workflows in practical healthcare environments.

What we learned

While building ShiftSaarthi, we learned a lot about workflow-focused product design, real-time speech interfaces, and AI-assisted operational systems.

We also learned how important usability and simplicity are in workflow software, especially for environments where users are already working under pressure.

What's next for ShiftSaarthi

Future improvements for ShiftSaarthi include:

-integration with hospital management systems and EHR platforms -secure cloud-based handoff storage -role-based authentication -mobile-first workflow support -improved multilingual medical vocabulary handling -analytics for workflow optimization across healthcare teams

The long-term vision is to make healthcare coordination faster, clearer, and more accessible through multilingual voice-first workflow automation.

Built With

• css
• googlegemini
• next.js
• speechapi
• tailwind
• typescript