In software industry there is a common assumption that deployment of software reliability engineering (SRE) contributes to huge overhead in development and its practice does not match the agile software development which puts emphasis on traveling light and generating a minimal amount of project artifacts. However measuring and/or assessing reliability of developed software is nevertheless a very important task and there seems to be a natural fit between SRE and agile in driving certification testing. SRE typically focuses on directing test efforts after software has been created in order to achieve a predetermined failure intensity objective (FIO). It also uses reliability growth models in order to predict when the software will meet the FIO, and therefore can be released. While these ideas may initially seem incompatible with the test-driven emphasis of agile methods, we show that SRE does in fact have value within the agile process context.