In February 2026, Ashafir developed a custom Databricks SQL dialect for the spiceai/spiceai repository, focusing on enhancing DataFusion’s compatibility with Spark SQL and Databricks SQL Warehouse. Leveraging Rust and SQL, Ashafir implemented the DatabricksDialect using DataFusion’s Dialect trait, enabling accurate translation of Spark-specific functions such as array_has and list_has, as well as support for backtick-quoted identifiers. Comprehensive tests were added to ensure the dialect’s reliability and regression safety. This work improved interoperability between DataFusion and Databricks, reducing integration friction and enabling seamless query translation across platforms, demonstrating depth in both software development and data engineering practices.