Inspired by the challenge of scaling up existing silicon quantum hardware, we investigate compilation strategies for sparsely-connected 2d qubit arrangements and propose a spin-qubit architecture with minimal compilation overhead. Our architecture is based on silicon nanowire split-gate transistors which can form finite 1d chains of spin-qubits and allow the execution of two-qubit operations such as Swap gates among neighbors. Adding to this, we describe a novel silicon junction which can couple up to four nanowires into 2d arrangements via spin shuttling and Swap operations. Given these hardware elements, we propose a modular sparse 2d spin-qubit architecture with unit cells consisting of diagonally-oriented squares with nanowires along the edges and junctions on the corners. We show that this architecture allows for compilation strategies which outperform the bestin-class compilation strategy for 1d chains, not only asymptotically, but also down to the minimal structure of a single square. The proposed architecture exhibits favorable scaling properties which allow for balancing the trade-off between compilation overhead and colocation of classical control electronics within each square by adjusting the length of the nanowires. An appealing feature of the proposed architecture is its manufacturability using complementary-metal-oxide-semiconductor (CMOS) fabrication processes. Finally, we note that our compilation strategies, while being inspired by spin-qubits, are equally valid for any other quantum processor with sparse 2d connectivity.