Stellar parameter determination represents a critical step in the procedure of Simple Stellar Population Synthesis (SSPS), of which the aim is to produce realistic composite spectra for model stellar populations, such as galaxies and star clusters. As SDSS-IV nears completion, the MaStar stellar library, an unprecedented large and comprehensive collection of medium-resolution stellar spectra from ~10,000 unique stars, will be made available to the public in its entirety. In order to make proper use of such a library in SSPS, accurate stellar parameter estimates (Teff, log g, [Fe/H], and [\alpha/Fe]) associated with the spectra are a necessity. Here, I present a new approach to stellar parameter estimation that we in the MaStar team have developed in recent years, which combines the spirit of conventional methods with the goal of obtaining an efficient and versatile software pipeline for delivering stellar parameter estimates, as well as other important data such as extinction estimates. This method makes use of the continuum-normalized stellar spectra and the spectral continua themselves (often characterized by photometric measurements in previous efforts) to match the MaStar spectra to models derived from the ATLAS9-based BOSZ model set via reduced-\chi^2 fitting. Carefully combining these two approaches allows us to avoid many of the problems that afflict each approach separately, and obtain reliable estimates throughout parameter space, including notoriously difficult regimes, such as at higher temperatures (> 7000 K). Steps are also taken to account for extinction and flux calibration residuals present in the data. Since this project is ongoing, focus will be placed on recent progress, preliminary results, and plans for the near future.