Optimizing linear algebra operations is critical for enhancing the performance of machine learning frameworks, which rely heavily on efficient matrix computations. High-performance libraries like Eigen play a pivotal role in accelerating these operations through architecture-specific optimizations. While Eigen is widely integrated into popular ML frameworks such as TensorFlow or PyTorch, its lack of support for emerging architectures like RISC-V limits its applicability in evolving hardware ecosystems. SHD Group projects that RISC-V processors will occupy 25% of the global market by 2030, underscoring the urgency of adapting foundational libraries to RISC-V. This work focuses on extending Eigen’s capabilities by implementing full support for RISC-V’s vector extension, addressing a key gap in its low-level optimizations. The enhanced Eigen library was benchmarked against OpenBLAS (pre-optimized for RVV) on three RISC-V platforms: Kendryte K230, Banana Pi, and Lichee Pi. For reference, performance was also measured on an ARM Cortex A73 board with NEON. The results demonstrate that Eigen’s RVV optimizations deliver a 2.5–3x speedup over scalar implementations and outperform OpenBLAS by 11% in key linear algebra operations. These improvements are particularly relevant for ML frameworks that depend on Eigen for critical computations, such as gradient descent, matrix factorization, and tensor contractions. By enabling efficient execution on RISC-V hardware, this work contributes to the growing ecosystem of energy-efficient and scalable ML solutions. The study highlights the importance of continuous optimization of core libraries to bridge the gap between emerging hardware architectures and the computational demands of modern machine learning.