Over the past ten years, we have identified star-forming galaxies at z>6, within the epoch of reionzation, using the Lyman break technique with deep fields from the Hubble Space Telescope and ground-based imaging. From the inferred rest-frame UV luminosity function, we can address the potential role of star formation in the reionization of the Universe. However, spectroscopic follow-up has been challenging, with Lyman-alpha (the main feature accessible from ground-based observations) often weak or absent. JWST will revolutionise this - NIRSpec works out to 5microns with huge multiplex, sampling the rest-frame optical out to z~6, and potentially obtaining redshifts from [OII] out to z~12 (and much higher with the Lyman break). I will describe the Guaranteed Observing Time programme with my colleagues on the NIRSpec instrument science team. Through emission lines and SED fitting, these observations have the potential to chart the evolution of the star formation rate, dust extinction, metallicity and stellar initial mass function within the first billion years.