The high mutation rate of Influenza A viruses (IAVs) necessitates yearly vaccine updates, and ongoing mutations in various animal reservoirs pose a significant pandemic risk. There is an urgent need to develop innovative therapeutic strategies with broad antiviral activity against diverse IAV strains. Structural analysis of domains within the IAV RNA polymerase, that are pivotal for viral replication, reveals high conservation across influenza subtypes. Targeting the critical sites on these conserved domains to impede viral replication offers a promising strategy to develop broad-spectrum therapeutics for influenza infections. Despite the significance of this protein, the development of effective small molecule inhibitors targeting it has not been remarkably successful to date. Computational protein design offers a promising tool kit to synthesize potent antiviral inhibitors to identify specific target sites amenable to inhibition. In this study, we will use deep learning-based de novo protein design methods to devise novel inhibitors of IAV replication. By leveraging RFDiffusion-ProteinMPNN-AlphaFold pipeline, we will generate a diverse repertoire of unique binders, each targeting a distinct conserved site on different subunits of the heterotrimeric IAV RNA polymerase. We anticipate that these de novo designed proteins will potently interfere with the replication activity of the virus. Unlike other molecules, that recognize a single epitope, mutations in the RNA polymerase (acquired through selection and mutation), should not lead to resistance affecting the inhibitor library, as each binder targets a distinct site. Hence they may serve as potent broad spectrum inhibitors of IAV replication and allow evaluating intracellular target sites suitable for antiviral intervention.