Thesis - Context Engineering for Autonomous AI Coding Agents

This thesis investigates whether well-engineered context can enable smaller or less capable language models to approach the performance of frontier models on realistic software development tasks. The role involves building a reusable context engineering pipeline to compare different context strategies, such as persistent project documentation, manual task-specific context, and automatically generated context. The evaluation focuses on measuring functional task success, performance gaps between model sizes, and the impact of context quantity and relevance on agent results.