Numba for custom algorithms.
Dataiku (amazing tool for preprocessing and complex flows)
Amazon RDS (postgress), but thinking about redshift.
Spark
Tableau or plotly/seaborn