Hand-held and in-line instrumentation based on short wave near infrared spectroscopy (SWNIRS; 400 – 100 nm) has been developed an applied to the assessment of fresh fruit maturity and eating quality. Attributes assessed include total soluble solids (TSS), total titratable acidity (TTA), dry matter content, skin pigmentation and flesh pigmentation. The use of chemometrics for wavelength selection, data pre-treatment and linear regression modelling for a range of internal quality attributes (TTA, TSS in fruit varying in starch content, firmness, internal flesh colour, maturity level, and internal defect’) of intact fruit is discussed, with optimisation of the wavelength range and the number of factors used in the Partial Least Square Regression model. The SWNIRS technique was demonstrated to be effective in the prediction of fruit flesh colour, dry matter (DM) and TSS content of a number of commodities (typical Root Mean Square of residual Errors of Prediction (RMSEP) around 1%). This technology has been commercially used to improve the eating quality of mango fruit in Australia.