A method for key predictors and machine learning for configuring battery cell performance may include providing a cell that includes a cathode, a separator, and a silicon-dominant anode; measuring a plurality of parameters of the cell; and using a machine learning model to determine cycle life based on the plurality of measured parameters, where one of the measured parameters includes second cycle coulombic efficiency. The plurality of parameters may include initial coulombic efficiency, cell impedance values, open-circuit voltage, cell thickness, and impedance after degassing. A first subset of the plurality of parameters may be measured before a formation process. A second subset of the plurality of parameters may be measured during a formation process, where the plurality of parameters may include a voltage reached during a first 10% of a first formation cycle. A third subset of the plurality of parameters may be measured during cycling of the cell.