The goal of the Ynot project is to make programming with dependent types practical for a modern programming language. In particular, we are extending the Coq proof assistant to make it possible to write higher-order, imperative and concurrent programs (in the style of Haskell) through a shallow embedding of Hoare Type Theory (HTT). HTT provides a clean separation between pure and effectful computations, makes it possible to formally specify and reason about effects, and is fully compositional. This seems like it's related to Adam Chlipala's A Certified Type-Preserving Compiler from Lambda Calculus to Assembly Language. See, in particular, slides 23-24 of this presentation (PDF). More generally, computation and reflection seem to be gaining recognition as important features for the practical use of Coq Again, the point is to simplify programming with dependent types in Coq
Russell is a language for programming with dependent types in Coq. It uses an adaptation of the predicate subtyping feature of PVS to allow users to write only algorithmic code while using strong specifications. Proof obligations are generated automatically, and, once proved, permit to build a complete, valid Coq term. As an example of using Russell to develop programs with dependent types, I implemented the Finger Tree data structure [3] in Coq. It gives quite a few insights about the power of dependent types for specification and their practical use [4]. Here's the relevant page. You can have a look at the example on celebrities in a party inspired by Richard Bird's article. Yet a lighter example: quicksort. I developed a complete formalization [5] of simply-typed lambda calculus with constants in the dependently-typed style with the help of Program. It includes a tait-style proof of weak normalization.