During April 2025, this developer focused on improving the correctness and reliability of scaled dot product attention computations in the tenstorrent/tt-metal repository. They addressed a critical issue in the decode path by fixing tensor dimension handling, ensuring buffer allocations matched actual tensor shapes. Using C++ and PyTorch, they updated the sdpa_decode program factory to accurately reflect these shapes, which reduced decode-path errors and improved operation accuracy. Their work included targeted memory allocation tuning and test harness adjustments, resulting in more robust and reliable attention computations. This contribution demonstrated a strong grasp of deep learning and low-level system reliability.