Identification of boosted heavy particles (i.e., top quarks or W, Z, Higgs bosons) from their hadronic decays can play an important role in searches for new physics at the LHC. We present a new approach for boosted jet tagging using particle-flow jets at CMS, where one dimensional convolutional neural networks are utilized to classify a jet directly from its reconstructed constituent particles. The new method shows significant improvement in performance compared to alternative multivariate methods using jet-level observables.