The distribution function is
$$F_x(x\mid\lambda) = e^{-e^{-\lambda x}}, \qquad \lambda > 0$$
Is this an exponential family?
The pdf that I obtained was
$$f(x\mid\lambda) = \lambda e^{-\lambda x} e^{-e^{-\lambda x}}$$
and the joint pdf of an iid sample is
$$f(x^n\mid\lambda) = \lambda^n e^{-\lambda \sum_{i=1}^n x_i} e^{-\sum_{i=1}^n (e^{-\lambda x_i})}$$
I don't see how I can bring this to the form that is required for it to be an exponential family. Am I right in concluding that this isn't an exponential family?
Also, is it possible to use the Karlin Rubin theorem to obtain a UMP test and thus a confidence interval for $\lambda$?