ABSTRACT:   Quantum computers process superpositions of information and will be necessary to tackle many classes of problems that are intractable using conventional computers. However, there are two big problems standing in the way: first, there are very few provable useful quantum applications; second, it is notoriously difficult to build quantum computers at scale. The highest performing (albeit small) quantum computers so far are composed of individual atoms as quantum bits (qubits), controlled and read out with laser beams. Such atomic quantum computers have allowed the abstraction of hardware for a new focus on algorithmic co-design instead of the physics of individual qubits. I will summarize the state-of-the-art in these quantum computers in both academic and industrial settings, for both highly programmable Hamiltonian quantum simulation and quantum circuit synthesis for applications in many areas of science. Hopefully this scientific path will develop into commercial utility, just as conventional computers in the 20th century.