This article highlights the inconsistency in output from AI coding assistants, such as Copilot, when standardization is left to individual prompting habits. It argues that the real problem in AI-assisted engineering is not capability, but rather the lack of consistency in code decisions.
Este conteúdo apresenta CGD-PD, uma camada leve para modelos de linguagem grandes (LLMs) que melhora a resposta a perguntas lógicas de três vias (Verdadeiro/Falso/Desconhecido). Ele aborda falhas recorrentes como inconsistência de negação e previsões 'Desconhecido' epistêmicas, utilizando decisões consistentes e desambiguação baseada em prova para maior precisão.
The content addresses the challenge of using AI systems, like LLMs, in production, where their probabilistic nature leads to inconsistent outputs, despite often being correct. The goal is to transform these inherently probabilistic systems to behave as consistently and repeatably as possible, bringing them closer to the determinism required for real-world workflows.
The author highlights common issues in AI agent workflows, such as stale states, incomplete handoffs, and unverified task completion. They introduce `agent-consistency`, an MIT-licensed Python package, to address these problems and seek feedback on its approach.