Android produces UI by producing and presenting a frame from a mobile app on the screen. If the app's UI rendering is slow, the system is forced to skip frames. When this occurs, the user notices a repeated flicker on their screen, which is known as Jank. Jank issues can manifest in various forms, causing disruptions such as unstable frame rates, heightened latency, AppDeadlineMissing, and BufferStuffing. BufferStuffing occurs when the app runs beyond its expected duration, leading to jank. To quantify this issue, we determine the total time taken by the app frame, commencing with the choreographer wake-up as the starting point and concluding with max(GPU, post time) as the endpoint. Post time represents when the frame was dispatched to SurfaceFlinger. Notably, due to the parallel operation of the GPU, the GPU can complete its task after the post time. This situation is more of a state than a true jank occurrence and typically arises when the app continually dispatches new frames to SurfaceFlinger before the previous frame has been presented. This continuous influx of frames leads to the stuffing of the internal Buffer Queue with frames yet to be presented, hence the term "Buffer Stuffing." These additional buffers in the queue are presented one after the other, resulting in increased latency. This can eventually reach a point where there are no more buffers available for the app to utilize, leading to a blocking wait during dequeuing. Importantly, even if the actual work performed by the app remains within the deadline, the stuffed nature of frames leads to their presentation at least one vsync late, introducing elevated input latency. While the visual appearance of frames may remain relatively smooth in this state, the late presentation is associated with increased input latency. In this work, the LSTM Model(Long Short Term Memory) was used for the detection of Jank. LSTMs provide us with a large range of parameters such as learning rates, and input and output biases. Hence, no need for fine adjustments. Experimental result shows that the LSTM Model was able to predict the frame drop with an efficiency of 98% thereby enhancing the overall user experience.