Build a matching algorithm that allows operators to query the proprietary DNA database, match compatible entities, and calculate the compatibility score.
We created the algorithm, following consultation with NRMP. We developed a fast elastic search, an operator interface, and API. This enabled integration with the customer’s various workflows.
- Working with DNA data of a suboptimal structure. We had to create a special module that replicated the remote DNA database in a necessary format to enable elastic search. Thus, we insured a fast search in spite of the weak original data structure;
- The use of the product is also enabled via API, and its usage has sporadic query number bursts. We built a queue layer for the queries that also had to store large amounts of query result data.
Platforms: AWS, EC2, S3, Elasticsearch Service