Inspiration
What it does
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for Untitled
Autopilot Agent – Your AI Coworker Inspiration
Modern professionals spend hours every day on repetitive administrative tasks such as reading emails, scheduling meetings, organizing to-do lists, summarizing documents, and preparing reports. We wanted to build an AI agent that doesn't just answer questions but actively helps users complete these tasks while keeping them in control through an approval workflow.
What it does
Autopilot Agent acts as an AI coworker. It can:
Prioritize and summarize emails. Generate reports from documents. Organize tasks and reminders. Schedule meetings. Create summaries of long files. Ask for user approval before executing important actions.
The goal is to reduce repetitive work so users can focus on higher-value decisions.
How we built it
We built the application using:
Qwen Cloud as the core large language model. Python for the backend. FastAPI for API development. Streamlit for the user interface. SQLite for storing user data and task history.
The application uses an agent workflow where user requests are analyzed, converted into tasks, and executed only after receiving user approval.
Challenges we ran into
One of the biggest challenges was designing an approval workflow that balanced automation with user control. Another challenge was enabling the agent to understand different task types while maintaining accurate and reliable responses. We also focused on creating a simple user experience despite multiple backend processes.
What we learned
This project taught us how to design AI agents that can perform actions instead of only generating text. We gained experience with prompt engineering, workflow orchestration, API integration, and building user-friendly AI applications.
What's next for Autopilot Agent
Future improvements include:
Integration with Gmail, Outlook, Slack, Google Calendar, and Microsoft Teams. Multi-agent collaboration for complex workflows. Long-term memory to personalize the assistant. Voice interaction. Enterprise security, permissions, and role-based access control. Deployment as a cloud-native productivity assistant for businesses.
Log in or sign up for Devpost to join the conversation.