Deep-learning systems and methods related to automated resolution of breaking changes in microservice-based software applications

A system for resolving breaking changes in a software system, the system comprising:

• a processor;
• and
• a memory storing instructions which, when executed by the processor, cause the processor to perform the steps of:
• generating, via an automated training data process, a training dataset comprising training examples, wherein, when described in relation to the generation of a first training example that is representative of how each of the training examples are generated, the automated training data process comprises:
• monitoring one or more communication channels to collect data associated with:
• test results of respective tests configured to identify breaking changes occurring in the software system;
• and management actions executed in the software system;
• storing test results data collected in relation to the test results in a timestamped historical test result database;
• storing management action data collected in relation to the executed management actions in a timestamped historical management action database;
• parsing content in the historical test result database and identifying failed test data associated with a first failed test of the test results;
• parsing content in the historical management actions database and identifying first management action data relevant to the first failed test, the first management action data comprising data describing the management actions taken between the first failed test and a next successful execution of the test of the first failed test;
• identifying the first failed test data and the first management action data as the first training example;
• training a machine learning model using the training examples included in the training dataset;
• using the trained machine learning model to predict one or more recommended management actions by:
• detecting a present failed test indicating a breaking change in the software system;
• determining failed test data associated with the present failed test;
• providing the failed test data to the trained machine learning model as an input and generating therewith the one or more recommended management actions as an output.