Utilizing machine learning models for automated software code modification

A device, comprising:

• one or more memories;
• and
• one or more processors, communicatively coupled to the one or more memories, configured to:
• receive requirement data identifying a requirement for modification of software code;
• process the requirement data, with a machine reasoning model, to identify one or more entities and one or more intents in the software code and to generate a query based on the one or more entities and the one or more intents;
• process the query, with a code locator model, to encode text of the query into high-dimensional vectors and to identify a semantic similarity between the high-dimensional vectors and code text provided in the software code,
• wherein the code locator model includes a universal sentence encoder that is trained with a two-deep averaging network encoder in parallel with a different drop rate, a composition function, and by determining an average of similarity confidence;
• process the query, the semantic similarity, and the code text, with a code developer model, to generate metadata;
• utilize the metadata to identify an identifier associated with the software code;
• determine, with the code developer model and utilizing the identifier, a portion of the software code;
• modify the portion of the software code based on the query to generate modified software code;
• and
• perform one or more actions based on the modified software code.

A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:

• one or more instructions that, when executed by one or more processors of a device, cause the device to:
• receive requirement data identifying a requirement for modification of software code;
• process the requirement data, with a machine learning model, to identify one or more entities and one or more intents in the software code and to generate a query based on the one or more entities and the one or more intents;
• process the query, with a code locator model, to encode text of the query into high-dimensional vectors and to identify a semantic similarity between the high-dimensional vectors and code text provided in the software code,
• wherein the code locator model includes a universal sentence encoder that is trained with a two-deep averaging network encoder in parallel with a different drop rate, a composition function, and by determining an average of similarity confidence;
• process the query, the semantic similarity, and the code text, with a code developer model, to generate metadata;
• utilize the metadata to identify an identifier associated with the software code;
• determine, with the code developer model and utilizing the identifier, a portion of the software code;
• modify the portion of the software code based on the query to generate modified software code;
• and
• cause the modified software code to be implemented in production.