The field of engineering and scientific research is witnessing a significant shift towards autonomous AI-driven discovery, with large language models (LLMs) playing a crucial role in this transformation. Researchers are developing innovative frameworks that leverage LLMs to automate complex workflows, such as finite element analysis, database auto-tuning, and storage system configuration. These frameworks are enabling rapid exploration of vast parameter spaces, leading to breakthroughs in various domains, including operations research, machine learning, and classical mathematical problems. Notably, the integration of LLM-based reasoning with large-scale evolutionary search is yielding state-of-the-art results in multiple areas. The trend towards autonomous AI-driven discovery is expected to accelerate innovation, automate complex discovery processes, and deliver substantial engineering and scientific advances with broader societal impact. Noteworthy papers include: FeaGPT, which introduces a fully integrated geometry-mesh-simulation-analysis pipeline for finite element analysis. Centrum, a novel model-based DBMS auto-tuner that improves distribution-free point and interval estimation in surrogate modeling. StorageXTuner, an LLM agent-driven auto-tuning framework for heterogeneous storage engines that reaches up to 575% higher throughput and reduces p99 latency by as much as 88%. The FM Agent, a general-purpose multi-agent framework that leverages a synergistic combination of LLM-based reasoning and large-scale evolutionary search to address complex real-world challenges.