Inspiration
We were inspired by how outdated traditional calendars feel in today’s AI-driven world. People still manually plan every detail, even though life is dynamic and constantly changing. We wanted to rethink personal time-management and build a system that adapts to users instead of forcing users to adapt to it.
What it does
Our project is an AI-powered personal management ecosystem that combines:
a Smart Calendar,
an AI Assistant,
an AI Friend,
and a Facts Generator for personalized daily insights.
It helps users plan their day, track habits, receive suggestions, resolve conflicts, and get short personalized “microfacts” about their routines. It transforms planning from a manual task into an intelligent, interactive experience.
How we built it
We built the system as an integrated AI platform combining:
a dynamic smart calendar with automatic scheduling logic,
LLM-based components for assistant and friend roles,
contextual data processing (habits, patterns, location, weather),
a micro-insight engine that generates personalized facts from user data.
The project consists of both frontend and backend modules, AI pipelines, and internal logic for context-aware recommendations.
Challenges we ran into
Designing a calendar that feels alive and not just automated reminders.
Combining productivity-focused AI with a more conversational “friend” AI in a way that feels natural.
Handling context awareness: location, weather, habits, routines, mood signals.
Transforming everyday data into small, meaningful insights.
Ensuring the system stays helpful without overwhelming the user.
Accomplishments that we're proud of
Building a Smart Calendar that can intelligently adjust, prioritize, and suggest optimal scheduling.
Creating two AI personas — Assistant and Friend — that work together instead of competing.
Implementing a Facts Generator that turns raw data into personalized micro-insights.
Designing an ecosystem that feels human, dynamic, and genuinely helpful.
Making everyday planning simpler, smoother, and more enjoyable.
What we learned
People don’t just need reminders — they need intelligent support that adapts to real life.
Context makes AI significantly more useful: weather, mood, location, habits all matter.
Small insights (microfacts, streaks, patterns) greatly increase user engagement.
Building “human-like” AI requires balancing personality, usefulness, and precision.
What's next for NULL-DTIT25
Adding achievement badges such as “50th lecture” or “1 year at a location”.
Expanding the Facts Generator to deeper behavioral analytics.
Improving natural voice input to make interaction even more fluid.
Exploring more adaptive personalization based on long-term trends.
Preparing the system for real-world deployment and scaling.
Built With
- ai
- api
- django
- fastapi
- newsql
- openai
- postgresql
- python
- react
Log in or sign up for Devpost to join the conversation.