Dual-capacity machine learning based validation framework

A method comprising:

• receiving, by a validation framework, target code generated by a code generator;
• receiving, by the validation framework, contextual information comprising at least a plurality of evaluation criteria and a reference evaluation;
• prompting by the validation framework, a large language model with the target code, the contextual information, the reference evaluation, and instructions to automatically evaluate the target code by performing operations comprising:
• segmenting the target code into a plurality of target code chunks, automatically evaluating each target code chunk of the plurality of target code chunks using a corresponding evaluation criteria of the plurality of evaluation criteria and the reference evaluation to yield a score of a plurality of scores,
• generating feedback for each target code chunk of the plurality of target code chunks, and
• determining a metric for the target code based on the plurality of scores,
• wherein the large language model, when prompted, performs the operations to output the metric and the feedback;
• and
• displaying, by the validation framework, the feedback for each target code chunk when the metric is less than a predetermined threshold.

A system comprising:

• a computer processor;
• a data repository in communication with the computer processor, wherein the data repository stores:
• target code having a plurality of target code chunks,
• contextual information comprising at least a plurality of evaluation criteria and a reference evaluation,
• instructions,
• a plurality of scores,
• feedback, and
• a metric;
• a large language model, when executed by the computer processor, performs operations to output the metric and the feedback, the operations comprising:
• segmenting the target code into the plurality of target code chunks,
• automatically evaluating each target code chunk of the plurality of target code chunks using a corresponding evaluation criteria of the plurality of evaluation criteria and the reference evaluation to yield a score of the plurality of scores,
• generating the feedback for each target code chunk of the plurality of target code chunks, and
• determining the metric for the target code based on the plurality of scores;
• and
• a validation framework, when executed by the computer processor:
• receives the target code,
• prompts the large language model with the target code, the contextual information, the reference evaluation, and the instructions to automatically evaluate the target code by performing operations, and
• displays the feedback for each target code chunk when the metric is less than a predetermined threshold.