Register now After registration you will be able to apply for this opportunity online.
AI-Driven Python Code Improvement
This project explores the application of Artificial Intelligence (AI) in enhancing Python code quality. It includes a literature review of traditional and AI-driven refactoring methods and analyzing existing tools and techniques for code improvements. Additionally, the project evaluates whether AI-assisted improvement maintains code correctness and reliability.
Keywords: AI, Code refactoring, Python, Large Language Models, Automation, Dataset creation
In this project, you support the young ETH Spin-Off “nihito” on their mission to empower everyone to build software for everyone.
This project explores the use of Large Language Models (LLMs) to automate and enhance Python code improvement. Traditional refactoring relies on rule-based techniques, whereas LLMs offer context-aware and adaptive transformations.
The study begins with a literature review, comparing existing classical and AI-driven refactoring methods, and evaluating existing tools for error detection, error correction, and code optimization.
A key focus of the study is developing AI agents powered by LLMs to automate Python code refactoring, including tasks such as function extraction and dead code removal. Furthermore, the agent must be able to detect code errors and provide suggestions for improvement.
To ensure reliability, the project investigates whether LLM-assisted improvement preserves code correctness through rigorous testing.
In addition to the technical aspects, this project offers a unique opportunity to gain insight into the workings of a young software start-up, giving you exposure to the fast-paced environment of an ETH Spin-Off.
In this project, you support the young ETH Spin-Off “nihito” on their mission to empower everyone to build software for everyone. This project explores the use of Large Language Models (LLMs) to automate and enhance Python code improvement. Traditional refactoring relies on rule-based techniques, whereas LLMs offer context-aware and adaptive transformations. The study begins with a literature review, comparing existing classical and AI-driven refactoring methods, and evaluating existing tools for error detection, error correction, and code optimization. A key focus of the study is developing AI agents powered by LLMs to automate Python code refactoring, including tasks such as function extraction and dead code removal. Furthermore, the agent must be able to detect code errors and provide suggestions for improvement.
To ensure reliability, the project investigates whether LLM-assisted improvement preserves code correctness through rigorous testing. In addition to the technical aspects, this project offers a unique opportunity to gain insight into the workings of a young software start-up, giving you exposure to the fast-paced environment of an ETH Spin-Off.
-Analyze and compare existing tools to determine the effectiveness of classical and AI-based code improvement methods.
-Develop AI agents to automate Python code improvement.
-Investigate whether AI-assisted refactoring preserves code correctness.
-Analyze and compare existing tools to determine the effectiveness of classical and AI-based code improvement methods. -Develop AI agents to automate Python code improvement. -Investigate whether AI-assisted refactoring preserves code correctness.