From Helpful to Trustworthy: LLM Agents for Pair Programming
For developers using LLM-based coding agents, this work addresses the problem of outputs being plausible yet misaligned with intent and lacking evidence for review, but the contribution is incremental as it proposes a study plan without concrete results.
This doctoral research proposes a systematic study of multi-agent LLM pair programming that externalizes developer intent and uses development tools for iterative validation, aiming to make LLM-generated code more reliable, auditable, and maintainable. The expected outcome is a clearer understanding of when multi-agent workflows increase trust and practical guidance for building reliable programming assistants.
LLM-based coding agents are increasingly used to generate code, tests, and documentation. Still, their outputs can be plausible yet misaligned with developer intent and provide limited evidence for review in evolving projects. This limits our understanding of how to structure LLM pair-programming workflows so that artifacts remain reliable, auditable, and maintainable over time. To address this gap, this doctoral research proposes a systematic study of multi-agent LLM pair programming that externalizes intent and uses development tools for iterative validation. The plan includes three studies: translating informal problem statements into standards aligned requirements and formal specifications; refining tests and implementations using automated feedback, such as solver-backed counterexamples; and supporting maintenance tasks, including refactoring, API migrations, and documentation updates, while preserving validated behavior. The expected outcome is a clearer understanding of when multi-agent workflows increase trust, along with practical guidance for building reliable programming assistants for real-world development.