We describe a non-parametric approach for accurate determination of the slowest relaxation eigenvectors of molecular dynamics. The approach is blind as it uses no system specific information. In particular, it does not require a functional form with many parameters to closely approximate eigenvectors, e.g., a neural network, and thus no extensive expertise with the system. The power of the approach is illustrated on long atomistic protein folding trajectories. The determined eigenvectors pass a stringent validation test at timescale of 0.2 ns, much shorter than alternative approaches.