Github - https://github.com/brcolli/devdayhackathon We have it running on our local host, do stop by at Table 5 to see (or call Ravi at 3124797402, we can come to desk)
Inspiration: We have been exploring small and medium-sized businesses (SMBs) where Generative AI and feature flags can add value. Personalized marketing is what we zeroed in on. One of the businesses where we saw a significant value proposition was the local honey business. We observed ample scope for creating custom experiences in the buyer's journey of purchasing honey, believing in creating a win-win experience for both buyers and sellers.
MyHoney: Most people use honey for a myriad of different reasons, such as health benefits, culinary uses, or as a natural sweetener. We aim to provide the best type of honey catered to each buyer's specific needs. Through personalized marketing, MyHoney ensures that customers receive recommendations for honey products that best suit their preferences and needs.
How We Built It: We utilized the following technologies: AWS and AWS Bedrock: For scalable infrastructure and AI model deployment. Python: For backend development and integrating AI models. React: For building a responsive and interactive frontend interface. LaunchDarkly: For implementing feature flags to manage and deploy features gradually, allowing for personalized experiences.
Challenges We Ran Into: Data Integration: Ensuring seamless integration of various data sources to create a comprehensive buyer profile. AI Model Training: Training AI models to accurately predict and recommend the best honey products for different user preferences. Feature Flag Management: Efficiently managing feature flags to provide a dynamic and personalized user experience without disrupting the existing functionality.
Accomplishments That We're Proud Of:
Successfully implemented a dynamic marketing platform that uses AI to personalize the shopping experience. Created a user-friendly interface that guides customers through a customized journey of discovering and purchasing honey. Developed a scalable solution that can be adapted and extended to other SMBs and industries.
What We Learned: The potential of feature flags extends far beyond simple feature toggles. They can be used to create highly personalized and dynamic user experiences. The importance of understanding user preferences and behavior to deliver truly personalized recommendations. The value of integrating various technologies to build a robust and scalable solution.
What's Next for MyHoney:
We plan to identify SMBs in the Bay Area that can benefit from this personalized marketing model. Offer a beta version of our application to these businesses for testing and feedback. Continuously improve our AI models and feature management system based on user feedback and data insights. Explore additional markets and applications for our dynamic marketing platform, extending beyond the honey industry.
Built With
- amazon-web-services
- awsbedrock
- launchdarkly
- python
- react
Log in or sign up for Devpost to join the conversation.