Unit testing is an important step in the software development workflow to detect bugs and ensure system correctness. Recently, Large Language Models (LLMs) have been explored to automate unit test generation and have demonstrated promising results. However, the generated tests are not always reliable, as they may contain syntax errors, hallucinations, test smells, or failing assertions. We conjecture that providing developers with feedback on such issues will increase the adoption of LLMs in realworld workflows. To address this, we propose PyTestGuard, a PyCharm plugin that allows developers to generate and refine unit tests directly within the Integrated Development Environment (IDE). Beyond test generation, PyTestGuard helps users evaluate test quality by detecting test smells and reporting issues such as missing arguments or references to non-existing objects. We conducted a user study with nine participants to assess PyTestGuard’s usefulness as a testing assistant and to identify areas for improvement. Participants reported that the tool’s feedback on test quality, along with its summarised error messages and coverage information, supported them while writing unit tests. However, they also faced challenges and suggested improvements before completely trusting LLM-based test generation in their development workflow. Based on these findings, we highlight several design recommendations for future tools that aim to integrate LLMs into software testing workflows.

Machine learning (ML) algorithms have been used frequently in the past years for Software Engineering tasks.
One of the popular tasks researchers use is method name prediction, which helps them generate an identifier for methods with ML models such as Code2Seq.
This model represents code snippets as Abstract Syntax Trees (AST) and includes all code elements except comments.
This paper introduces a novel approach to incorporating comments in Code2Seq.
The reimplemented models integrate comments into the AST when preprocessing the data.
We filtered comments to reduce the noise introduced in the model and improve the predictions using techniques such as TF-IDF and stopword removal.
We implement several models for each filtering technique and one for raw comments.
These models are trained and evaluated using the dataset java-small, provided by Code2Seq, with 700K examples and the following metrics: precision, recall, and F1.
The results obtained from the evaluation of our model with raw comments show an improvement of 3.62% and 2.36% in precision and F1, respectively, on method name prediction compared to the original model.
Furthermore, the model with stopword removal has a 6% and 3.52% gain in recall and F1. These improvements suggest that adding comments to Code2Seq is valuable for better method name predictions since they contain additional information about the methods.