HomeAI

Simplify home buying with HomeAI. Get clear, unbiased info fast, minus the legal jargon and forms.

Comment

Inspiration

Jenny is a first-time home buyer who lacks the competence and confidence necessary for navigating home disclosure and other critical parts of making a house her first home…

What it does

Home.AI serves as a reliable and unbiased platform equipped with extensive knowledge and sophisticated communication features. It is designed to revolutionize the home buying experience by assisting users in navigating through property disclosures and offers, ensuring a transparent and informed purchasing process.

How we built it

  • User and Expert Interviews: Initiated our design process by interviewing target users (home buyers) and real estate agents to identify pain points and gain insights.
  • Data Management: Imported real estate disclosure data into AstraDB, utilizing it as our vector database for efficient data handling.
  • Information Extraction: Employed LlamaParser to extract clean information from PDFs, ensuring data quality and accessibility.
  • Framework Integration: Used LlamaIndex to seamlessly integrate GPT-4 with AstraDB, forming the backbone of our system's architecture.
  • Prompt Engineering: Enhanced both the quality of input and output through meticulous prompt engineering in the backend, optimizing user interactions.

Challenges we ran into

  • Data Complexity: Encountered properties with disclosures exceeding 100 pages of unstructured data, requiring effective parsing and indexing for usability.
  • Desire for Structured Outputs: Transitioned from conversational to structured outputs via LLM or backend processing to provide users with clear, intuitive property disclosures. Instead of conversation style outputs, we want structured outputs (by either LLM or backend processing). So that users receive clean data on their interested properties, and have clean, intuitive disclosure analysis.
  • Prompt Engineering: Invested in prompt engineering to ensure clean, formatted data presentation.
  • Technical Hurdles with LlamaIndex: Faced frequent "token exceeds limit" issues, suggestive of potential bugs. Overcame these through customized query engine prompts, enabling a workaround.

Accomplishments that we're proud of

  • User-Centric Design: Prioritized user needs by empathetically addressing pain points through thoughtfully designed features.
  • Effective Technology Use: Leveraged RAG and LlamaParser for efficient data extraction and storage, ensuring clean and structured outputs.
  • Business Potential: Recognized the significant commercial opportunity our project presents. SOM is 19m, SAM is 1.9B
  • Originality: Inspired by our real-life experiences, our project stands out due to its uniqueness and relevancy.

What we learned

  • Impact of Home.ai: Recognized Home.ai's potential to significantly save time and money for users in their home purchasing journey.
  • Technical Proficiency: Gained a deep understanding of RAG, vector databases, LlamaIndex, and LlamaParser, boosting our confidence in utilizing these technologies for application development.
  • Importance of Communication: Learned the value of seeking assistance and the importance of proactive communication when encountering obstacles.

What's next for Home.AI

  • Comprehensive Buyer Journey Support: Expand Home.AI to offer end-to-end assistance, covering every phase of the buyer's journey—from initial research to smart price suggestions.
  • Negotiation Assistance: Integrate tools to aid in the negotiation process, empowering buyers with data-driven support.
  • Technician Coordination: Implement features to connect buyers with technicians for accurate home repair cost estimates, streamlining the post-purchase process.

Built With

+ 10 more
Share this project:

Updates