This serves as a de facto storage for our prompt engineering and dataset retrieval from ChatGPT. The methodology is RHLF: where Afraz, Rayan & Mukarram have reviewed the outcomes of the dataset.
Format: Text files, will load to JSON storage later.
Model Used: GPT 4o
Prompt Used: Consider yourself as my pair dataset curator. We are going to generate a dataset that will redefine software engineering interviews. Let’s take {paradigm} for example. We will have a conversational interview, and you will generate three responses for each question. These ‘answers’ will score from 0 to 10 (10 being the highest) based on 0 to 10. Please keep in mind that I will also appreciate you as a candidate being able to relate to real-world examples (although this will be for the 10 one). {any paradigm-specific conditions go here, examples are in SQL query generation, etc). Please generate the output in a JSON form, so it is ready to feed in an LLM. I’ll ask the questions, you generate the answer.
Concepts Covered: Arrays, two pointers, Stack, Binary Tree, Dynamic Programming, Binary Heap