Developed and delivered QwQ-32B API inference support for the HuanzhiMao/gorilla repository, focusing on backend development and model deployment using Python. The work introduced a centralized API inference handler for the Qwen API, streamlining integration and reducing overhead for downstream services. By configuring the QwQ-32B model within a unified API surface, the implementation improved accessibility and established a foundation for future inference capabilities. All changes were committed with clear traceability, ensuring production readiness and facilitating future enhancements. The project demonstrated skills in API integration, model configuration, and version control, contributing to a more consistent and maintainable inference pipeline.