Publications

Test-Driven Prompting: An Empirical Evaluation of Methods Reconstruction based on Unit Tests

• Event: CBSoft 2026 (XVII Congresso Brasileiro de Software: Teoria e Prática). Presented there; the presentation was only for people whose paper was accepted for publication.
• Status: accepted, waiting for the journal publication.
• What it evaluates: how much using unit tests as context for the AI affects code generation.
• Result: in general the results are good, which matches what we see in the day-to-day practice. It let me connect the scientific side with my daily practice, bringing explanations backed by data.
• Co-authors: João Eduardo Montandon de Araujo Filho
• Certificate: CBSoft 2026 - Certificado de Apresentação