Additive manufacturing (AM), specifically laser powder bed fusion (LPBF), holds considerable significance across diverse industries such as tool and die making, IC manufacturing, medical implants, electronic cooling, and aerospace sectors due to its ability to produce components with intricate internal and external geometries. This stands in contrast to traditional manufacturing methods that impose limitations on the complexity of part designs. Notably, there is a growing interest in using AM for the production of components with intricate cooling channels featuring complex surface topographies designed to enhance thermal performance. While conventional machining methods can address the roughness of external surfaces, they fall short when it comes to treating internal channels, especially when these dimensions are at a millimeter or submillimeter scale. AM emerges as a promising alternative with the potential to overcome this limitation. To ensure the successful industrial adoption of AM for parts requiring sophisticated cooling channels, it becomes imperative to comprehend the relationship between the as-built surface finish and heat transfer. In LPBF, numerous build parameters, such as part orientation during the build, significantly impact the final part surface topography and, consequently, heat transfer. Existing literature, exemplified by Moody's diagram, simplifies the treatment of surface roughness. However, powder bed fusion processes generate intricate surfaces characterized by strong anisotropic features, spatter, and surface defects, all of which have the potential to influence heat transfer and fluid flow. This research primarily focuses on investigating the effects of AM roughness characteristics, including scan orientations, density of spatter deposits, sizes of spatter, amplitudes, wavelengths, etc., on heat transfer from corresponding AM surfaces and pressure drop across cooling channels. The study employs both numerical and experimental approaches. Computational Fluid Dynamics (CFD) models for mini-channels using StarCCM+ were developed, integrating roughness data from real AM surfaces. CFD simulations for the entire system model and modeling of mini-channels with different wavy surfaces aided in establishing suitable dimensions for experimental setups and determining the Reynolds numbers necessary for relevant experiments. An exchangeable experimental setup was developed based on CFD findings, and AM parts with critical scan orientations (0°, 45°, and 90°) were fabricated, subsequently machined to fit into the setup. In addition, a smooth-surfaced Inconel part served as the baseline control condition. Comparative analysis of CFD and experimental results across different Reynolds numbers validated the findings, revealing significant differences in Nusselt numbers and pressure drops among various AM surfaces. The surface with a 90° scan orientation demonstrated superior heat transfer performance based on nominal build conditions. Building upon these outcomes, further investigation into the effects of 90° weld tracked surfaces in a circular form was conducted. Two aluminum (Al-6061) channels—one with a smooth surface and the other with internal threads mimicking artificial waviness similar to an AM surface with a 90° scan orientation to the fluid flow direction—were conventionally manufactured. Both CFD and experimental investigations were conducted for different mass flow rates. The results indicated that artificial waviness had a substantial impact on heat transfer, resulting in high cooling efficiency. The Nusselt number was approximately three times larger for various flow conditions compared to the smooth channel. However, intentionally structured surfaces also led to larger pressure drops, potentially necessitating additional pumping power depending on the specific application.