Video Agents and Embed Enhancements for the Agentic Web — May 21, 2026

The primary goal of my workday was to improve functionality for video agents as well as improving the embeds throughout the site. Completing all the tasks took a total of 9 hours. Some of the major features launched include a lead capture popup for video agents, enhanced functionalities for embeds, and other improvements to Tavus video integration. These changes are necessary, as they streamline user experience and increase overall agentic web platform user functionality.

An example of my work is to create a lead capture popup that is intended to capture user information while they are interacting with the video agents. This functionality adds to the overall user focused interactive experience. I dedicated a significant portion of my time making sure that the newly added feature is in sync with the previously created systems so that it won’t interrupt user interactions.

Some of the other tasks I performed include critical fixes to the connection of embeds and videos. With these fixes the the embed drawer is now available with a complete panel of Portal Details that allows users to control their experience with the platform better. I encountered some difficult problems that were created with the permissions setting of the iframe in regards to video embeds. The error messages the system was displaying were not clear, however I was able to configure the proper settings related to our X-Embed-Host-Origin after a thorough investigation. This fix was also very important to make sure that the video embeds operated properly and were able to work with external sites.

An improvement was made with the Tavus video integration, which now includes controls for the microphone and a ‘Leave’ button. These controls provide more video call control to users as they can now manage their audio. Getting the microphone permissions to work in the iframe was a challenge, but after several tries, I got it to work. This is a feature that will help a lot with the usability of video calls.

I also worked on the security of the MapLibre integration of the platform. I made some security hardening measures and improved the user experience in call hangups. These changes are important for the security of the platform and the user experience. I spent a lot of time testing to ensure there were no bugs and that everything worked.

It is clear that building this platform on my own with the AI tools like Claude Code in VS Code has its pros and cons. While I sometimes miss having a larger team to brainstorm with, the freedom lets me iterate quickly. I get help from the AI for coding and debugging, but the actual implementation and testing takes time. I really spent a lot of time on this today as I went through bugs and issues that made the day longer than I expected.

I am making progress, as each shipment is another feature I thought to build, and each one is an improvement to my dream of starting a one-man-show company valued at a billion dollars. The platform boosts my vision of an agentic web. Every feature provides lasting value, and I measure success in increments so each step is an indicator of success. Today was monumental in expanding our video capabilities.

To summarize, I spent my day improving our video agent and embed user experience, fixing several bugs, and making sure our platform is robust enough to sustain continued growth and usability.

I'm excited about what's next to come and how it'll change some of our user engagement and satisfaction metrics. Building this platform is a marathon, not a sprint. Each day presents new problems and new areas for improvement. I am confident I will continue to improve.