This paper proposes a dynamic buffer occupancy correction architecture for streaming applications using the VVC video coding standard. The solution employs reinforcement learning (QLearning) to adapt the video bitrate according to buffer status, aiming to optimize the user’s Quality of Experience (QoE). Experimental results show significant reductions in rebuffering events, improved buffer stability, and increased objective video quality when compared to static threshold-based approaches. The proposed method is evaluated under adverse network conditions with varying bandwidth and latency.