In this chapter, we examine the role of emerging energy brokers in a local energy market (LEM), facilitating indirect local energy trading. This chapter proposed a reinforcement learning-based local energy trading strategy that aims to diversify the energy ecosystem at the edge of distribution networks rather than replace existing energy services or become the best trading model. The trading mechanism provides additional options for customers and prosumers to participate occasionally in the retail electricity market, supplementing existing utility services. Considering customer behavior characteristics, it improves trading efficiency and encourages local power balance. The energy trading process will be structured as a Markov decision process (MDP), incorporating reinforcement learning and data-driven methods. Meanwhile, economic concepts like search friction, related to typical search costs in this trading model, are also discussed.