• 

  home page

LifeOS

Inspiration

As a student, I often find myself managing assignments, project deadlines, internship responsibilities, appointments, and personal commitments across multiple apps. It can be overwhelming to keep track of everything, and there have been times when I spent more time deciding what to do next than actually doing it. I wanted a system that could remember my commitments, help me prioritize my responsibilities, and guide me on what to focus on each day. This inspired me to build LifeOS, an AI-powered personal operating system designed to help people stay organized and productive.

What It Does

LifeOS is an intelligent life administration agent that helps users manage their daily responsibilities through natural conversation. Users can tell the agent about deadlines, appointments, goals, and tasks, and LifeOS will organize them, prioritize them based on urgency and importance, and generate actionable plans.

Using MongoDB as a persistent memory layer, LifeOS remembers important information across sessions, allowing it to provide personalized recommendations and maintain context over time.

How I Built It

I built LifeOS using Gemini for reasoning and planning, MongoDB Atlas as the long-term memory database, MongoDB Vector Search for retrieving relevant memories, and the MongoDB MCP Server to allow the agent to interact directly with stored data. The system is orchestrated through Google Cloud Agent Builder and presented through a simple web interface.

Challenges I Faced

One of the biggest challenges was moving beyond a simple chatbot and creating an agent that could actually remember information and use it effectively. I also had to design a memory system that could retrieve the most relevant tasks and commitments at the right time while keeping the experience simple and intuitive for users.

What I Learned

Through this project, I learned how agentic AI systems differ from traditional chatbots and how persistent memory can significantly improve the usefulness of AI assistants. I also gained hands-on experience with MongoDB's vector search capabilities and learned how AI agents can combine reasoning, memory, and planning to help users take meaningful action rather than simply provide answers.

Built With

• gemini