Enhancing the thermal performance of heat exchangers while minimizing flow resistance is critical for efficient energy systems. This study focuses on the effect of varying blade counts (2, 4, 6, and 8 blades) and Reynolds numbers (6000, 8000, 10,000, 12,000, and 14,000) on the thermal performance factor (TPF), friction factor, and heat transfer coefficient in thermal boundary layer heat exchangers. Numerical analyses were conducted to generate comprehensive datasets capturing the influence of blade-induced swirl and turbulence intensification across the specified Reynolds number range. An Artificial Neural Network (ANN) was developed and trained using these datasets to predict thermal and hydraulic performance with high accuracy. The model was used to optimize design parameters, identifying configurations that maximize heat transfer enhancement while controlling frictional penalties. Results indicate that increasing blade count significantly improves the heat transfer coefficient and TPF, especially at higher Reynolds numbers, but also leads to a corresponding increase in friction factor. The ANN demonstrated excellent predictive capability, offering a reliable tool for design optimization and performance prediction in thermal boundary layer heat exchangers with high Reynolds numbers and high blade counts. This work provides valuable insights into the interplay between blade geometry and flow conditions and demonstrates the effectiveness of AI-assisted modeling for the next generation of compact, high-performance heat exchangers.