Ananya Amancherla developed the initial integration of the VitisAI Execution Provider for the microsoft/onnxruntime-genai repository, enabling ONNX Runtime to leverage hardware-accelerated inference on Vitis AI-capable devices. She designed and implemented configuration options in C++ to register a custom operations library, laying the foundation for hardware-optimized inference workflows. Her work focused on API design and software architecture, establishing the necessary infrastructure for future hardware validation and continuous integration testing. By addressing the requirements for enterprise-grade performance and reduced latency in GenAI workloads, Ananya’s contributions provided a robust starting point for ongoing hardware acceleration support within the project.