Inspiration
Our inspiration for ScanBack stemmed from the incredible world of extreme coupon saving, where people save significant amounts of money on their purchases. As we marveled at the TV shows and stories about individuals who mastered the art of couponing, we asked ourselves a simple question: "Why couldn't we create an app that harnesses this power for everyone?"
What it does
ScanBack is a revolutionary app that empowers users to optimize their shopping experiences at Walmart and similar grocery stores. Our app allows users to effortlessly scan their Walmart receipts and highlights areas where online stores offer better deals on the same products. By identifying these opportunities, users can take full advantage of Walmart's price-matching policy to save money. ScanBack brings the world of extreme couponing to the digital age, helping users find hidden discounts and ensuring they never overpay for products available at a lower cost elsewhere.
How we built it
BrightData's remote webproxy to avoid getting rate limited while scraping websites for price data
For the backbone of our application, we developed a robust machine learning model that can accurately identify potential savings opportunities on scanned receipts. We utilized Python and state-of-the-art machine learning libraries to create this model, ensuring it could adapt to various receipt formats and quality levels.
Challenges we ran into
Developing ScanBack presented several challenges, with one of the most significant hurdles being the creation of our machine learning model. Teaching the model to accurately identify savings opportunities, especially when dealing with distorted or damaged receipts, required extensive testing and fine-tuning. We invested a considerable amount of time and effort into enhancing its accuracy and robustness.
Additionally, ensuring the app works seamlessly and efficiently for users on different devices and platforms was another challenge we encountered. Compatibility and user experience were top priorities, which led to iterative design and development phases.
Accomplishments that we're proud of
Despite the challenges we faced, we're immensely proud of the accomplishments we achieved during the development of ScanBack. Notably, we crafted a sleek and user-friendly frontend that enhances the overall experience. Our backend code is not only effective but also idiomatic, making it easier for future developers to understand and maintain our application.
We successfully created a machine learning model that can analyze Walmart receipts and identify savings opportunities accurately, providing users with valuable information that can help them save money on their purchases.
What we learned
Throughout the development of ScanBack, we learned a great deal about the power of machine learning in the context of consumer savings. We gained a deep understanding of the challenges involved in processing and interpreting various receipt formats and how to optimize the model for better performance.
Furthermore, we honed our frontend development skills, focusing on creating an intuitive user interface that is both aesthetically pleasing and highly functional.
What's next for ScanBack The future of ScanBack is exciting. We have several enhancements and features in the pipeline. These include:
Expanding Retailers: We plan to extend our service to cover other major retailers, allowing users to find the best deals across a broader spectrum of stores.
User Profiles: Implementing user profiles, tracking savings over time, and providing personalized shopping recommendations.
Community Features: Building a community where users can share their saving success stories and tips, fostering a sense of belonging among our users.
With these enhancements, ScanBack aims to revolutionize the way people shop, helping them save money effortlessly and consistently.
Log in or sign up for Devpost to join the conversation.