Jisen Li developed a Web Agent Demo for the NexaAI/nexa-sdk repository, showcasing local execution of complex web tasks using the Qwen3-VL multimodal model. The project integrated the Nexa SDK with a user-facing web UI to enable autonomous browser automation, demonstrating end-to-end vision-language understanding. Jisen focused on demo setup, documentation, and web automation, providing clear setup instructions and usage examples to streamline onboarding and rapid prototyping. The work, implemented primarily in Bash and Markdown, addressed the need for accessible demonstrations of advanced web automation capabilities, delivering a well-documented, functional feature without bug fixes during the development period.