- Complete the whole end to end query generation pipeline - Add the latest bert classifier - Integrate the column retrieval (column pruning) with the newly created pipeline #4 - Completly integrate each components