Abstract: Parallel computers are growing in capability and complexity. Their potential for societal impact is also increasing in wide range of areas including biomolecular modeling for understanding disease and drug design, genetics/bio-informatics climate and weather prediction, design of engineered artifacts and materials, as well as data analytics and machine learning in various forms. A particular formulation of a programming model we pioneered helps improve productivity and automates many runtime performance optimizations. We have been working on these in concert with development of multiple interdisciplinary application codes – for biophysics, astronomy, materials and so on. I will present a broad overview and current snapshot of key challenges and ongoing work.