Inspiration
As a first-year computing student, I'm always juggling assignments and deadlines, and opening an app to type them in is a hassle. I wanted to manage my study schedule by simply talking, the way you would to a voice assistant like Alexa+.
What it does
StudyBuddy is a simulated Alexa+ study planner. You speak or type to it, and it can: Add assignments with due dates, e.g. "add Python quiz due Friday" List what's due, e.g. "what's due" Mark work as done or delete it Build a study plan with a one-hour session the day before each deadline, e.g. "plan my week" It replies out loud and in text, and shows each tool call it makes on screen.
How we built it
It's a single-page web app in HTML, CSS and JavaScript, with no frameworks or dependencies. The browser's Web Speech API handles voice input and spoken replies. A rule-based intent parser turns each sentence into a tool call (add_task, list_tasks, complete_task, delete_task, plan_week), which mirrors the tool-calling pattern Alexa+ uses with MCP integrations. Assignments are saved in the browser's localStorage. The code is open source under the MIT licence.
Challenges i ran into
Understanding natural dates like "Friday", "tomorrow" and "in 3 days" without a library, and handling local time zones correctly. Writing rules flexible enough to understand different ways of saying the same thing. Speech recognition isn't available in every browser, so typing works as a fallback. Building and submitting the whole project on a tight deadline alongside my studies.
Accomplishments that i am proud of
A working voice-driven assistant with no external dependencies. Clean, readable code that others can fork and extend. Publishing my project as open source for the first time in a hackathon.
What i learned
I learned how voice assistants map spoken requests to tools, how the Web Speech API works, and how to handle dates reliably in JavaScript. I also learned how to scope a project so it can be finished and submitted on time.
What's next for StudyBuddy for Alexa+
Turn the simulated tools into a real MCP server so it can run on Alexa+. Use a language model for more flexible understanding of what people say. Add reminders, recurring tasks and calendar sync. Add support for more languages, including South African ones.
Log in or sign up for Devpost to join the conversation.