Antibody engineering technologies face increasing demands for speed, reliability and scale. We developed CeVICA, a cell-free antibody engineering platform that integrates a novel generation method and design for camelid heavy-chain antibody VHH domain-based synthetic libraries, optimized in vitro selection based on ribosome display and a computational pipeline for binder prediction based on CDR-directed clustering. We applied CeVICA to engineer antibodies against the Receptor Binding Domain (RBD) of the SARS-CoV-2 spike proteins and identified >800 predicted binder families. Among 14 experimentally-tested binders, 6 showed inhibition of pseudotyped virus infection. Antibody affinity maturation further increased binding affinity and potency of inhibition. Additionally, the unique capability of CeVICA for efficient and comprehensive binder prediction allowed retrospective validation of the fitness of our synthetic VHH library design and revealed direction for future refinement. CeVICA offers an integrated solution to rapid generation of divergent synthetic antibodies with tunable affinities in vitro and may serve as the basis for automated and highly parallel antibody generation.