The search for photoabsorbers, materials that can harness solar photons and transform their energy into electrical charge, has attracted much attention from researchers working in fields of materials science, chemistry, biology, and physics. This resulted in a huge amount of theoretical and experimental efforts to understand mechanisms, predict properties, synthesize new materials, and characterize them, all in search of yet more efficient, cheaper, and environmentally friendly compounds and assemblies. Recently, thanks to the development of computational resources and experiment automation, focus has shifted towards trial and error approaches. In theoretical studies, these are comprised of high-throughput screenings, where small numbers of compounds with preferential properties are sieved out from a larger family of materials based on simple descriptors, using fast ab initio atomic scale simulation methods. The proposed project aims to complement material searches with comprehensive studies based on state-of-the-art high-level of theory methods, developed for higher precision, but computationally more demanding. The main tasks would be to work on including all relevant physical phenomena into models and develop modified descriptors for material searches. Focus will be put especially on the nature of the charge carriers and their interaction with defects close to the surface, and in ambient conditions. To model band structures of the materials in an accurate way, GW approximation will be used. Interactions between positive and negative photogenerated charges will be assessed within the BSE method. Hybrid functionals will be used to study energetics and structure of polarons and defects. Temperature effects on the electronic structure and optical properties will be studied through molecular dynamics simulations. We will investigate materials to be used in solar cells, like halide perovskites and kesterites, and water-splitting cells, such as vanadates or tungstates.