HerRestartAI

Inspiration

Every end is the start of a new beginning. We wanted to build something meaningful for single mothers, women navigating divorce, and students who are struggling to understand how to rebuild financially. Many financial tools assume stability. They assume steady income, shared expenses, and predictable costs. But that is not reality for everyone. Financial transitions can feel overwhelming, isolating, and uncertain.

We wanted to create something that feels supportive during those moments. Something that provides clarity instead of confusion and structure instead of stress. Instead of guessing what expenses might look like, we looked at real-world U.S. economic data to understand how housing, food, transportation, and other essential costs have actually shifted in recent years. We wanted the guidance to reflect economic reality, not assumptions.

What It Does

HerRestartAI analyzes a user’s current financial situation and evaluates their overall stability. Based on income, dependents, housing costs, and debt, it generates a realistic and structured financial growth plan tailored to that specific situation. The system produces a personalized monthly budget breakdown, a projected debt-free timeline, actionable steps for the next 30 days, and a structured six-month recovery plan. It also generates a clear AI-written summary that explains the reasoning behind the recommendations in a supportive and easy-to-understand tone.

Because financial planning can feel intimidating, the platform also reads the plan aloud using AI voice technology. This makes the experience feel more like speaking with a financial assistant rather than interacting with a static budgeting tool. The goal is not just to calculate numbers, but to build confidence.

It can also read the plan out loud using AI voice, making it feel more like a financial assistant rather than just a budgeting tool.

How We Built It

On the frontend, we used JavaScript, HTML, and CSS to create a dynamic and responsive interface that updates in real time as soon as a user’s plan is generated. On the backend, we built the logic in Python and integrated the Google Gemini API to generate personalized financial plans. We also integrated the ElevenLabs Text-to-Speech API to convert the AI summary into natural-sounding voice output.

We structured user inputs into a financial profile and enriched our understanding using real U.S. expense data to better reflect current cost-of-living patterns. This helped us design prompts that were grounded in actual economic conditions rather than hypothetical scenarios. The backend processes the structured profile, sends it to Gemini for plan generation, and returns a detailed, organized response that the frontend renders into a clean dashboard experience.

Challenges We Ran Into

One of the biggest challenges was designing prompts that produced structured, actionable financial guidance instead of generic advice. Financial recommendations must feel specific and realistic, especially when users are already in vulnerable situations. We also faced Gemini API rate limit constraints and had to carefully optimize requests to ensure reliability during development.

Choosing the right voice tone for ElevenLabs was another important decision. The voice needed to feel supportive and calm rather than robotic or overly formal. On the technical side, integrating frontend and backend systems under hackathon time pressure required careful debugging. Parsing natural voice responses also presented challenges, especially when handling unexpected answers like “none” or incomplete phrases. Working with real-world expense data required cleaning and interpreting datasets to extract meaningful insights that could inform our design.

Accomplishments That We’re Proud Of

We successfully built a fully functional end-to-end system within hackathon constraints. The application integrates multiple APIs seamlessly and delivers structured financial plans in real time. The voice-based interaction adds a human-centered layer to the experience, allowing both voice input and output. The dashboard updates dynamically and presents financial insights in a clean, intuitive way. Most importantly, the recommendations are informed by real-world economic trends, making the system more grounded and realistic.

What We Learned

This project deepened our understanding of backend and frontend integration under time pressure. We learned how to work with multiple APIs simultaneously and how to design prompts that generate structured financial outputs rather than vague summaries. We gained experience handling real-world edge cases in voice input and debugging live integration issues quickly. We also learned how public economic data can be interpreted and used to inform AI-driven recommendations.

What’s Next for HerRestartAI

The next phase includes building a more advanced financial stability scoring system that dynamically adjusts recommendations as user conditions change. We plan to incorporate more detailed budgeting logic and scenario planning, such as modeling income drops or unexpected expenses. Future development will also include beginner-friendly investment education for women who are new to financial markets, secure user accounts for long-term tracking, and a progress dashboard that visualizes financial growth over time. We also aim to integrate live economic indicators so recommendations can adapt to changing cost-of-living conditions.

Built With

• api
• css
• elevanlabs
• gemini
• html
• javascript
• python