Inspiration

Our journey began with a desire to push the boundaries of what our current AI agent, Finn, could do. While Finn was already a competent and reliable assistant, we saw potential for it to become faster, more context-aware, and better at understanding nuanced instructions. Inspired by the rapid advancements in natural language understanding and user-agent interaction, we set out to design an upgraded version—Finn on steroids.

What it does

With Finn on steroids, we’ve taken a significant step toward building a more intelligent, helpful, and personable AI agent. While there's always room to grow, we’re proud of what we’ve accomplished—and excited for what's next.

How we built it

Frontend: React, TypeScript, Tailwind CSS
Backend: Node.js, Express
Database: PostgreSQL with Drizzle ORM
AI Integration: NVIDIA AI API
Speech Recognition: Web Speech API
Deployment: AWS

Challenges we ran into

The importance of multi-modal context integration (combining text, code, and visual inputs). Finding the perfect idea.

Accomplishments that we're proud of

Seeing how the vague vision we had at the start came to a standing product.

What we learned

How to build an end-to-end application. Work productively with AI agents by prompt engineering. Working with bunq API.

What's next for BunqHackathon - Finn on steroids

Refining the product so it can be deployed in production. Adding more feature to make Finn stronger and customer experience easier.

Share this project:

Updates