Modernizing information systems is a recurring need for large enterprises. Data migration from source to target information system is a critical step for successful modernization project. Central to data migration is \emph{data transform} that transforms the source system data into target system. Though there are different commercial tools available to address data migration challenge, creation of data transformation specification is largely a manual, knowledge intensive, and expert driven process. In this paper we present a tool that assists the experts while creating the data transformation specification by suggesting candidate field matches between the source and target data models and rules for the data transformation. Our tool is adaptive in the sense that it can take the user feedback in terms of corrected matches and validation data, and then proposes new matches and transformation rules for the remaining fields. Our tool uses machine learning, knowledge representation in order to learn and infer the candidate matches and it uses program synthesis to infer the transformation rules. We have executed our tool on real industrial data. Our schema matching recall at 5 score is 0.76, which means the experts need to look into first 5 tool recommended matches to identify the correct field match for 76 out of 100 fields. The recall at 2 score of the rule generator is 0.81, which means the experts need to look into first 2 tool suggested transformation rules to identify correct rule.