Abstract Mud pulse telemetry is a critical technology in the domain of oil and gas drilling operations, facilitating the real-time transmission of data acquired by downhole sensors to the surface. This study presents a case study focusing on an AI-aided surface detection system specifically developed for mud pulse telemetry, with the objective of enhancing data rates, accuracy, and overall reliability. The system utilizes transmission channel response information for pulse detection, effectively mitigating the adverse effects of reflected pulses that frequently impede high data rate transmission. Empirical results obtained from a field trial unequivocally demonstrate a substantial improvement in effective throughput when compared to conventional correlation-based detection systems.