Project Info

HomeEase

Devpost

Home maintenance and decor often require navigating multiple apps — one for plumbing, another for interior design, and yet another for product shopping. We wanted to create a one-stop, intelligent home platform that helps users solve any home-related problem — whether it’s fixing a broken faucet, repainting a wall, or visualizing a new decor idea — all powered by AI automation and smart agents. Our goal was to make home management as easy as asking a question — speak, type, or upload an image, and let HomeEase handle the rest. HomeEase provides AI-driven assistance for decor, renovation, and home maintenance. Users interact through voice, text, or images, and the system intelligently creates a job request. They can choose between two main paths: DIY Mode — The Fetch.ai agent analyzes the request and generates: A step-by-step repair or renovation plan A list of required tools and products Amazon links (scraped via Bright Data) for easy purchase Service Mode — Another Fetch.ai agent locates nearby service providers for the job. For home decor, users upload an image of their room. The system suggests products, scrapes Amazon, and then uses another Fetch.ai image-generation agent to merge the product visuals into the original image, helping the user visualize how the decor would look. Frontend: React with TailwindCSS for a clean, responsive user experience Backend: FastAPI, python Voice Processing: LiveKit for real-time voice input and transcription AI Agents: Multiple Fetch.ai agents for: Generating DIY steps and tool lists Finding local service providers Combining decor images with product visuals Web Scraping: Bright Data to scrape Amazon for recommended products Integration: Each agent’s output flows into the next stage, forming an autonomous task pipeline Image Processing: A generative model integrates the decor products into the uploaded image Agent Coordination: Ensuring smooth data flow between multiple Fetch.ai agents with asynchronous tasks. Voice-to-Text Accuracy: Maintaining reliable speech recognition and transcript interpretation with LiveKit. Web Scraping Reliability: Handling rate limits and dynamic pages while scraping Amazon product listings. Image Blending: Combining input images with product visuals while maintaining realistic composition. Context Management: Preserving job context across DIY and Service modes without user re-entry. Built an end-to-end AI workflow combining voice, image, and agent-based automation. Successfully integrated LiveKit, Fetch.ai, and Bright Data within a single project pipeline. Achieved realistic decor visualization that gives users a tangible preview of their design ideas. Enabled a seamless transition from AI-driven DIY to local professional assistance. How to coordinate multiple autonomous AI agents for distinct subtasks while maintaining state consistency. Best practices for real-time voice input handling and integrating third-party APIs efficiently. The importance of user experience in AI-driven apps — clear feedback and interactivity make all the difference. Ethical and practical considerations when using web scraping and AI-generated imagery. Integrate budget estimations and time-to-complete predictions. Add payment and scheduling features for service bookings. Enable multi-agent collaboration for larger renovation projects. Include personalized style recommendations using computer vision. Launch a mobile version of HomeEase for on-the-go interactions.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • CSSIn code
  • HTMLIn code
  • JavaScriptIn code
  • PythonIn code
  • ReactIn code
  • SQLIn code
  • SupabaseIn code
  • Tailwind CSSIn code
  • TypeScriptIn code

9 of 9 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

527 KB

Source files

127

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars