Inspiration
Managing personal finances is overwhelming for many - especially students, young professionals, and small business owners. Budgeting apps today mostly show raw numbers without offering real, contextual insight. We wanted to create a tool that doesn’t just track your spending, but helps you understand it. Our inspiration came from the idea of empowering people to make confident financial decisions with personalized AI support.
What it does
Funq is a smart financial assistant powered by Agentic AI. It helps users make informed decisions by:
- Analyzing past spending habits
- Forecasting the impact of future purchases
- Comparing financial behavior to others in similar demographics
- Answering general financial questions in a conversational format
How we built it
Funq is built on top of the bunq API, enabling access to users’ transaction history and balances. We implemented four cooperating AI agents:
- Orchestrator: Determines which agents are needed and reformulates user prompts.
- Future Financial Assistant: Evaluates future purchases using web search and user data.
- Past Financial Advisor: Analyzes spending patterns and compares them to peers.
- General Assistant: Handles all other user queries.
These agents collaborate to provide clear, actionable advice within a chat interface and a financial summary dashboard.
Challenges we ran into
- Designing smooth coordination between multiple AI agents
- Ensuring secure and accurate access to financial data
- Creating meaningful demographic comparisons while respecting user privacy
- Balancing detailed financial analysis with a friendly, intuitive user experience
Accomplishments that we're proud of
- Successfully implemented multi-agent orchestration with real-time user interaction
- Delivered meaningful, personalized financial insights—not just raw numbers
- Created a clean, user-friendly chat and dashboard interface
- Integrated external web data with internal transaction history for context-aware advice
What we learned
- How to orchestrate and align multiple LLM-based agents for a single task
- The importance of prompt engineering in improving AI output quality
- The value of user context and personalization in financial tools
- Technical integration with financial APIs like bunq's
What's next for Funq
We plan to:
- Add more features by extending the number of tools and agents Funq has access to
Log in or sign up for Devpost to join the conversation.