MARCO: Multi-Agent Code Optimization with Real-Time Knowledge Integration for High-Performance Computing

--Large language models (LLMs) have transformed software development through code generation capabilities, yet their effectiveness for high-performance computing (HPC) remains limited. HPC code requires specialized optimizations for parallelism, memory efficiency, and architecture-specific considerations that general-purpose LLMs often overlook. We present MARCO (Multi-Agent Reactive Code Optimizer), a novel framework that enhances LLM-generated code for HPC through a specialized multi-agent architecture. MARCO employs separate agents for code generation and performance evaluation, connected by a feedback loop that progressively refines optimizations. A key innovation is MARCO's web-search component that retrieves real-time optimization techniques from recent conference proceedings and research publications, bridging the knowledge gap in pre-trained LLMs. Our extensive evaluation on the LeetCode 75 problem set demonstrates that MARCO achieves a 14.6% average runtime reduction compared to Claude 3.5 Sonnet alone, while the integration of the web-search component yields a 30.9% performance improvement over the base MARCO system. These results highlight the potential of multi-agent systems to address the specialized requirements of high-performance code generation, offering a cost-effective alternative to domain-specific model fine-tuning. High-performance computing (HPC) represents the pinnacle of computational power, utilizing clusters of computing resources to overcome the limitations of individual machines. HPC's fundamental advantage lies in implementing parallel processing techniques that maximize processor cluster performance, enabling complex data processing and mathematical calculations that would otherwise be infeasible [33]. HPC has been instrumental in driving innovation across diverse domains including climate modeling, astrophysics simulations, pharmaceutical research, energy optimization, financial risk analysis, and training state-of-the-art machine learning models, particularly Large Language Models [7, 11, 15, 20, 30, 47].