You are collaborating with a team of researchers to develop cutting-edge algorithms for financial analysis. Your team frequently develops and debugs intricate TensorFlow models. You aim to balance the ease of debugging with the need to reduce model training time. Considering these requirements, how should you configure your training environment?