first cut at readme update including more information about HiOpBBpy

Author: thartland

README.md

@@ -123,11 +123,23 @@ HiOp supports three types of optimization problems, each with a separate input f
 
 More information on the HiOp interfaces are [here](src/Interface/README.md).
 
+# HiOpBBpy - python based black-box optimization
 
-## Issues
-Users are highly encouraged to report any issues they found when using HiOp.
+HiOpBBpy is a black-box (derivative free) Bayesian optimization solver for solving certain mathematical optimization problems. HiOpBBpy primarily targets problems for which the objective is very expensive and derivative information is not available. HiOpBBpy leverages MPI parallelism for batched Bayesian optimization methods and other embarassingly parallel tasks in said methods.
+
+## Build/install instructions
+
+HiOpBBpy can be built via pip. A standard build can be done by invoking the following from the hiop home directory
+```shell 
+$> pip install .
+```
+
+## Dependencies
 
+HiOpBBpy dependencies are listed in the pyproject.toml file.
 
+## Issues
+Users are highly encouraged to report any issues they found when using HiOp.
 
 ## Acknowledgments