Two-loop corrections to the Lamb shift

We present a new calculation of the order alpha^2 (Z alpha)^5 correction to the Lamb shift, which uses methods developed for multi-loop calculations.