BeforeYouSign

Inspiration

BeforeYouSign was inspired by how difficult it can be for regular people to understand lease agreements before signing them. Residential leases often contain important details about fees, deadlines, repairs, utilities, renewal terms, deposits, and tenant responsibilities, but those details are usually buried inside dense legal language.

That creates a real problem for students, first-time renters, international students, and people who cannot afford professional lease review. Many renters sign because they feel rushed, not because they fully understand what they are agreeing to.

I wanted to build a tool that helps renters slow down, understand the document, and know what questions to ask before signing.

What it does

BeforeYouSign is an AI-powered lease review tool that makes residential leases easier to understand before signing.

Users can upload a lease PDF, paste lease text, or try a sample lease. The app analyzes the lease and produces a clear report with:

• A plain-English summary
• Important lease terms
• Potential red flags
• Fees and money-related obligations
• Deadlines and notice requirements
• Tenant and landlord responsibilities
• Questions the renter should ask before signing
• Evidence quotes from the lease text
• A risk band that helps users understand how concerning the lease may be

BeforeYouSign does not replace a lawyer. It acts as a first-pass review tool that helps renters become more informed before they commit to a lease.

How we built it

BeforeYouSign was built as a web application using Next.js, React, TypeScript, and Tailwind CSS.

The app supports both PDF upload and pasted lease text. For uploaded PDFs, the system extracts the lease text, normalizes the content, and sends it through a lease-analysis pipeline. The analysis combines rule-based checks with AI-generated explanations so the output is both structured and readable.

The report is organized into sections so users do not just get a large wall of text. Instead, they can quickly see the most important issues, including money terms, deadlines, responsibilities, and red flags.

A major design goal was grounding. When the app identifies a concern, it tries to connect the issue back to evidence from the lease. This helps users understand why the issue matters and where it came from.

The risk scoring is intentionally simple and transparent. Rather than pretending to give legal certainty, the app uses risk levels to help users prioritize what they should review more carefully.

Challenges we ran into

One challenge was making the app useful without making it overconfident. Lease review can be sensitive because users may make real decisions based on the output. Because of that, BeforeYouSign had to be framed as an educational and decision-support tool, not as legal advice.

Another challenge was grounding the analysis in the actual lease text. It is easy for an AI-generated report to sound helpful while being too generic. I had to focus on making the output tied to specific lease evidence, especially for red flags.

PDF handling was also a challenge. Lease documents can be formatted in many different ways, and extracted text can be messy. The app needed to handle uploaded files, pasted text, and sample leases while still producing a consistent report.

The final challenge was keeping the user experience simple. A lease can contain a lot of information, but the app needed to show the most important points without overwhelming the user.

Accomplishments that we're proud of

I am proud that BeforeYouSign solves a practical problem that many people can understand immediately. It is not just an AI demo. It helps users make sense of a real document before making a serious decision.

I am also proud of the report structure. The app does not only summarize the lease. It separates the output into useful categories like fees, deadlines, responsibilities, red flags, and follow-up questions.

Another accomplishment is the focus on transparency. The app avoids presenting the result as a final legal judgment. Instead, it helps the renter understand what may need more attention before signing.

I am proud that the project is accessible to the type of user it is meant to help: someone who may not know legal terms, may be reviewing a lease quickly, and needs a clearer explanation before making a decision.

What we learned

I learned that building with AI is not just about getting a model to generate text. The harder part is designing the system around the model so the output is useful, structured, and trustworthy.

I also learned the importance of narrowing the problem. “Analyze any legal document” is too broad. Focusing on residential leases made the product clearer, easier to test, and more useful for a specific audience.

Another lesson was that good user experience matters as much as the model response. Renters do not need a huge legal essay. They need clear priorities, plain language, and specific questions they can ask.

I also learned that risk scoring should be used carefully. A simple score can help users prioritize, but it should not pretend to replace professional judgment.

What's next for BeforeYouSign

Next, I want to improve BeforeYouSign by making the analysis more specific and more reliable across different lease formats.

Future improvements include:

• Better clause detection for rent, deposits, late fees, utilities, renewal terms, and maintenance responsibilities
• Stronger evidence highlighting so users can jump from a red flag to the exact lease text
• More sample leases for testing and demos
• A clearer exportable report that renters can save or bring to a housing office, landlord, or advisor
• Support for comparing lease versions before and after changes
• Better guidance for what questions to ask before signing

The long-term goal is to make BeforeYouSign a practical renter-support tool that helps people understand what they are agreeing to before they sign.

Built With

• gemini
• nextjs
• typescript
• vercel