Imagine AI effortlessly streamlining complex engineering projects, optimizing everything from airplane designs to intricate brake systems. That's the tantalizing promise of new research exploring the power of Large Language Models (LLMs) for combinatorial optimization. Traditionally, optimizing complex systems like aircraft or automobiles relies on intricate mathematical models and algorithms. But these methods often struggle to capture the nuanced real-world dependencies inherent in engineering design. This is where LLMs step in. Researchers have developed a novel framework that uses LLMs to optimize Design Structure Matrices (DSMs), which visually map the interdependencies of tasks or components in a project. The framework feeds the LLM both the mathematical representation of the DSM and contextual domain knowledge, such as the names and descriptions of the components. The LLM then suggests an optimal sequence of design activities or parameter adjustments to minimize feedback loops and streamline development. In experiments with real-world DSMs of an Unmanned Combat Aerial Vehicle, a microfilm cartridge, a heat exchanger, and an automobile brake system, the LLM-driven approach significantly outperformed traditional optimization methods. It achieved faster convergence to optimal solutions, meaning it found good designs more quickly. Interestingly, providing the LLM with domain knowledge further boosted its performance, showing that LLMs can combine semantic understanding with mathematical reasoning to tackle complex engineering challenges. While promising, there's still room for improvement. Future research will explore larger, more intricate networks and test the framework on a broader range of optimization problems. The ability to analyze the LLM's intermediate reasoning steps is also a key goal, offering valuable insights into its decision-making process. This research opens exciting possibilities for LLMs to revolutionize engineering design, offering a powerful new tool to tackle the increasing complexity of modern systems.