Task: Intel Labs had a clear demo story and needed help with the technical development and implementation of the demo. The setup involved three servers with different configurations, and the challenge was to build one dashboard that would connect with them simultaneously, view and interact with a web UI, and receive and graph the data.

Process: Automating webpages was something new to me, so it was a research and learning process. The RAG chatbots were built with LangChan and Chainlit and I needed a way to interact with multiple Chanlit chat UIs at once.

After a few prototypes, I found a solution using TouchDesigner to hold the overall demo UI and load the webpages, then simulate user interaction with both pages simultaneousy with Python. I created a single text input field that would pass characters to each webpage, and python functions for more complex tasks such as moving a slider to a set point.

I worked with the engineering team to create a quick http sharing protocol with the servers, so that the demo could receive live performance data.

Impact: The demo clearly demonstrates the value of SVS in a live, competitive environment. The demo was featured at Intel Innovation 2024, and Google Cloud Next 2025.