Coding assistants like GitHub Copilot can speed up development, but their suggestions often miss important project context. This limits how well they align with a team’s conventions, tools, and domain knowledge. Keeping this context accurate and up to date is challenging, especially as codebases evolve. In this assignment, you explore how project context can be automatically maintained and used to improve Copilot’s suggestions in a practical, scalable way.
Software developers increasingly rely on coding assistants like GitHub Copilot for code generation and suggestions. Copilot, however, operates with limited project context, often leading to suboptimal code completions. Enhancing the contextual input it receives could significantly improve its usefulness in real-world development environments.
GitHub Copilot generates code based on the current file and surrounding code, but lacks sufficient global project context to make accurate or efficient suggestions. For instance, when asked to write unit tests, Copilot does not always use the project’s preferred unit testing libraries. It also does not understand domain-specific terms. One solution involves creating a dedicated file with “Copilot instructions” to provide this context. However, writing and maintaining this file is time-consuming and subject to documentation drift, especially in large or rapidly evolving codebases.
Furthermore, the instructions file should be detailed enough to guide Copilot effectively, but concise enough to remain readable and maintainable and to fit well within the LLM’s context window. As the codebase evolves, this file must be updated regularly. In early project stages, rapid changes are expected, whereas later stages require more stability.
The Assignment
Find a way to automatically and periodically update a context file based on the evolving state of a software project. The system could:
-
Analyze the structure and evolution of a codebase to identify relevant context updates
-
Implement an adaptive update mechanism that avoids excessive or unnecessary changes
-
Explore optimization strategies to balance stability and adaptability, such as simulated annealing
-
Evaluate the impact of the generated instructions on Copilot’s performance, quality, and usefulness through developer feedback and/or code suggestion metrics
About Info Support
Info Support specializes in custom software, data/AI solutions, management, and training and is active in the Finance, Industry, Agriculture, Food & Retail, Mobility & Public, and Healthcare sectors. We provide solid and innovative solutions for complex and critical software issues. Our headquarters are located in Veenendaal (NL) and Mechelen (BE). At present, approximately 500 employees are employed by Info Support.
Info Support’s working method is characterized by a number of core values: solidity, integrity, craftsmanship, and passion. These core values are intertwined in our work and the way we interact with each other.
To ensure that all employees are always up to date with the latest developments, Info Support has an in-house IT Academy that eagerly satisfies the hunger for more or different knowledge and skills.
B2 language proficiency in Dutch is required.
