During this meeting, the State Changers discussed the creation of an API with Xano. The purpose of the API is to suggest connections for a user based on shared keywords and shared connections. The discussion was focused on trying to make this process as efficient as possible.
Initial attempts included comparing two lists and seeing what overlaps as well as performing fuzzy searches on blocks of text. However, due to the extensive computation that may be involved and the potential slowness of these approaches, other more efficient methods were discussed. One alternative method mentioned was to have the data processing occur as background tasks with the results then cached in a table. This asynchronous approach separates the computational load temporally, making the endpoint response faster. The group also discussed making comparisons using arrays instead of block texts and exploring the use of 'tags' for exact matches. There was a proposed approach to separate out the queries and then compile the results after sorting. A point of interest during the meeting was creating these APIs in two distinct parts: the first part involves iterating over a model using a for loop to determine the length of overlap, and the second part involves sorting the output based on the overlap. The meeting concluded with an intention to test these combined APIs for performance, and continue this process in subsequent discussions. The group explored multiple strategies and showed a consistent focus on efficiency and speed. An advanced understanding of Xano was demonstrated throughout.
(Source: Office Hours 9/19 )
