-
Book Overview & Buying
-
Table Of Contents
Supercharged Coding with GenAI
By :
In Part 2 of the book, we explored LLMs in greater depth. We explained how they work, what they excel at, and how to leverage prompt engineering techniques to achieve more effective results. We also covered strategies for evaluating their outputs to ensure reliability.
This chapter took the concept of few-shot learning a step further by demonstrating how to fine-tune an LLM to specialize on a given task. Through positive and contrastive training examples, we guided the model to generate function implementations based solely on their signatures, returning clean code without inline comments. This approach can be applied more broadly to tasks such as generating unit test suites, maintaining docstring quality, or refactoring for loops across an entire repository.
With this deeper understanding of LLMs, prompt engineering, and output evaluation, we now have the essential tools to become supercharged coders. We can determine the best tool for a given task, whether ChatGPT...