To reduce the cost of integrated circuit fabrication, many industries are leveraging the fabless model, where the process of fabrication is distributed around the world. There are more entities involved in the fabrication process of design and a security breach can happen at any point such as in the form of a hardware trojan. The traditional approaches try to find the condition that can activate the trojan, and if the hardware trojans are present, the side channel signature (such as delay) will differ from the golden model. The traditional approaches use the Static Timing Analysis (STA) data that is obtained before the fabrication, however, the post-fabrication STA data can be different, which happens due to the sources of variability like; 1) voltage noise; 2) Process Variation (PV); and 3) process drift. The attacker, on the other hand, can cleverly embed the hardware trojan whose effect is masked by pessimistic margins due to account of sources of variability. The Neural Network Assisted Timing Profiling for Hardware Trojans Detection solution (NATE) proposed, aims to train a Neural Network to correlate the static timing data (produced at design time) to the delay information obtained from the clock frequency sweeping (at test time) for the purpose of Trojan detection. This can be used to detect the stealthy hardware trojans.