Cleaning validation is essential in pharmaceutical manufacturing to prevent cross-contamination, yet traditional wipe-sampling and chromatographic methods are labor-intensive and time-consuming. This study demonstrates handheld Fourier transform infrared (FTIR) spectroscopy combined with partial least squares (PLS) regression as a rapid, nondestructive alternative for quantifying naproxen surface residue on stainless steel. A total of 293 FTIR spectra were collected from stainless steel coupons spiked with naproxen (0 to 5.15 μg cm−2) over 650 cm−1–4000 cm−1. After outlier removal (IQR method), 263 spectra were retained and preprocessed using Savitzky–Golay smoothing and standard normal variate (SNV) normalization. The global PLS model (fivefold CV) achieved R2=0.854, RMSE = 0.491 μg cm−2, limit of detection (LOD) = 1.95 μg cm−2, and limit of quantification (LOQ)= 5.92 μg cm−2. Interval PLS identified the 1646–1810 cm−1 region (naproxen C = O stretch) as most predictive, yielding markedly improved performance: R2=0.973, RMSE = 0.211 μg cm−2, LOD = 0.716 μg cm−2, and LOQ = 2.169 μg cm−2. These results demonstrate that chemometric interval selection substantially enhances sensitivity and chemical interpretability, establishing handheld FTIR-interval partial least squares as an efficient, nondestructive alternative for routine pharmaceutical cleaning verification.