Program synthesis is a technique to find a program automatically from the given set of programs, i.e., a grammar, to satisfy some user-provided specifications. Significant research over the years has led to the evolution of program synthesis from synthesizing string manipulation programs in Microsoft Excel to the logic locking of digital circuits. Though program synthesis was designed to synthesize programs automatically, recent works in this area show that it can also be used in other domains.

In this manuscript, we are trying to address three problems where we use program synthesis techniques to solve them. In the first problem, we try synthesizing fences and atomic blocks in concurrent programs under relaxed memory models. The second work focuses on synthesizing semantic actions of attribute grammars. The last work is about synthesizing abstract transformers for an abstract domain to automate abstract interpretation.