Increasing power consumption from advanced computing and artificial intelligence (AI) systems has surpassed the capabilities of traditional air cooling due to rising heat output. This project explores direct-to-chip liquid cooling, specifically through microchannel-based cold plates as a more effective method for managing high thermal loads in densely packed systems. The study evaluated cold plates with varied microchannel geometries under controlled flow conditions. Measurements included temperature, flow rate, and pressure drop within a custom built closed-loop cooling test system. Additionally, a heat transfer model and calculator were developed for practical industrial applications. The project consists of design, testing, and refinement phases for a validated cooling solution and an accessible, web-based heat transfer calculator. This research aims to provide efficient and compact cooling solutions that are essential for scalable and sustainable AI and data center operations.