Fiscally

Inspiration

Fiscally was born out of my frustration with the complexity of tax filing. Many individuals and small business owners struggle with organizing tax documents, understanding deductions, and navigating tax laws. With the rise of AI-driven automation, I saw an opportunity to create a tool that simplifies tax management and empowers users to handle their taxes with confidence.

What it does

Through this project, I gained a deeper understanding of AI-driven financial applications, secure document handling, and user-centric design. I also honed my skills in backend development, API integrations, and cybersecurity best practices for handling sensitive financial data.

How I built it

Fiscally is built using a combination of React, Typescript, Tailwind CSS, Axios, and Auth0 for the front end and Node.js, and Mongo DB Atlas for the backend. I integrated the Google Gemini API to power the intelligent chat assistant, Taxi, ensuring that users receive personalized tax insights. For document uploads and storage, I implemented AWS S3 with encryption for enhanced security. The tax analysis engine leverages machine learning models to provide users with key insights into their finances.

Challenges I ran into

One of the biggest challenges was ensuring data security and compliance with financial regulations. Implementing encryption and secure authentication protocols required significant research and testing. Another challenge was optimizing the AI assistant, Taxi, to provide accurate and contextually relevant tax advice without overcomplicating the user experience. Balancing AI automation with human-like intuitiveness was an iterative process that involved fine-tuning prompts and incorporating real-time feedback.

Accomplishments that I'm proud of

I’m especially proud of successfully developing an AI-driven assistant that provides personalized tax insights, implementing secure document uploads with encryption, and designing a functional tax analysis engine. Overcoming technical challenges such as ensuring data security and compliance with financial regulations was a significant achievement. Additionally, seeing Fiscally evolve from an initial idea into a working application has been a rewarding milestone in my journey as a developer.

What I learned

Working on Fiscally deepened my understanding of AI-driven financial applications, secure data management, and user-focused design. I gained experience integrating Gen AI API to create an intelligent tax assistant, implementing secure authentication and encryption for sensitive financial data, and optimizing machine learning models for tax insights. Additionally, I learned how to balance automation with usability, ensuring that AI-powered features enhance rather than complicate the user experience.

What's next for Fiscaly

Moving forward, I plan to enhance Fiscally by improving Taxi’s accuracy in tax-related responses through more advanced AI fine-tuning and expanding the platform’s capabilities to support small business tax filings. I also want to integrate real-time tax law updates and provide users with proactive tax-saving recommendations. Additionally, improving the UI/UX for a more seamless and intuitive experience is a priority, along with exploring partnerships with financial institutions for greater credibility and reach.

Built With

• amazon-web-services
• auth0
• axios
• css
• dom
• gemini
• mongodb
• node.js
• react.js
• router
• sdk
• tailwind
• typescript