This presentation covers computational design of advanced composite materials and topology optimisation subject to additive manufacturing (AM) constraints. The first study examines the computational design and mechanical characterisation of Helical Carbon Fiber Composites (HCFCs). Traditional straight carbon fibre tows possess limited ductility and fracture resistance, which restrict their effectiveness in load-bearing and impact-sensitive applications. To address this shortcoming, a bio-inspired helical fibre architecture is proposed to improve both strength and toughness. A theoretical framework is established to evaluate the influence of helix angle on tensile response, supported by a newly developed intersecting circular cross-section model that captures the geometric effects of helical microstructures. Experimental testing shows that a helix angle of 5° significantly enhances tensile strength and fracture toughness. A geometry-based elastic modulus prediction model achieves an accuracy within 2.8%, outperforming conventional approaches, and is further verified through finite element simulations. In addition, a multiscale machine learning approach is introduced to relate mesoscale helical parameters to macroscopic mechanical behaviour through a structured optimisation workflow. The second study focuses on topology optimisation for additively manufactured structures. Key AM-related challenges, including enclosed cavities, overhang limitations, material anisotropy, and process-induced uncertainty, are addressed using an AM-oriented optimisation framework. This framework integrates self-support design, connectivity control, and robust topology optimisation within a bi-directional evolutionary structural optimisation (BESO) methodology. An anisotropic material description combined with a hybrid uncertainty model is employed to account for variability introduced by the AM process. Numerical examples demonstrate that both anisotropy and uncertainty have a pronounced impact on optimal structural layouts, underscoring the necessity of robust and manufacturability-aware optimisation strategies.