Additive manufacturing of fiber-reinforced composites via material extrusion enables the production of structurally optimized, anisotropic components tailored to specific load conditions. A key challenge, however, is that conventional topology optimization methods do not incorporate filament paths, and existing approaches for filament path planning often fail to efficiently support the structural function of already optimized geometries. To address these limitations, this work presents two contributions. First, a staggered optimization approach is introduced that couples two level-set methods to optimize topology and filament paths simultaneously, ensuring performance and manufacturability. Second, a hybrid three-step framework is proposed to generate fabrication-ready filament paths that align closely with local principal stress directions. The effectiveness of these methods is demonstrated on two-dimensional structures, where the optimized filament layouts achieve superior compliance compared to state-of-the-art approaches. Practical feasibility is further validated through multi-material 3D printing with variable-width deposition, which effectively eliminates gaps and preserves the intended mechanical performance. These results demonstrate the framework’s capability to bridge the gap between theoretical design optimality and real-world manufacturing constraints, providing a robust pathway for producing high-performance composite components via additive manufacturing.