Collaborating with engineers, I played a role in developing the interface for their recommendation engine, which made use of AI technology.
The idea was to upload multiple databases, allowing the tool to produce a personalized list of tire suggestions based on the particular car make and model selected by the customer using the fitment engine I had created.
Step 1 - Developing the Machine Learning Recommendation Tuner
This was a dynamic, hands-on collaboration between design and backend engineering. Together, we worked on developing a tool that engineers could utilize to upload databases and datasheets, enabling the machine to learn about establishing correlations and connections.
This involved not only determining which tires or wheels would be most suitable for various types of vehicles, but also offering suggestions for installers in particular locations.
I immediately immersed myself in analyzing the issue and the necessary specifications for this tool to address.
The tool needed to be capable to upload datasets and organize as well as being able to display and filter data in different tiers. The following designs are allowing to look at the data in various ways and allow to manually change the weight or correlation between product/brand, Product/Performance and Product/warranty
Being able to uplaod databases of product libraries was paramount
Once the databases had been uploaded and the correlations edited Engineer wanted to have the ability to test results and see the before and after tuning
Step2 - Developing tools to seamlessly integrate into the customer experience
The tuner and recommendation will smoothly blend with different parts of the customer journey, starting with an improved fitment engine. This engine will be present across the entire experience, including the Homepage, Landing page, List and detail pages, and notably on the recommendation popup or pages.
The fitment Engine
