This is great, I’m working on something that takes advantage of SVG also. Different use case but I do think there’s a lot unexplored value in SVG files.
Years and year ago, I worked in a sign makers and helped out with screen printing on occasion. Screens can last a very long time provided they are created corrected. From memory there were at least 200+ prints from a single screen - I think they may have been t-shirts or tote bags.
It was a code editor that had Gemini integrated into it. I’ve been playing around with the new version, it can still do what I was using it for, but it does make me wonder if it will become a OpenClaw like tool.
I have to wonder if helpful ads most likely refers to the type of add for the exact search result…
So, for example if you search for Coca-Cola the first ad will be something Coke has paid Google for. That helps Google earn $ and helps Coke not loose to a site with better SEO and confusion. Does it help you… maybe.
I’ve been building a small ‘agent’ using copilot at work, partly a learning exercise as well as testing it in a small use case.
My personal opinion is that AI and agents are being misrepresented… The amount of setup, guidance and testing that’s required to create smarter version of a form is insane.
At the moment my small test is:
Compressed instructions (to fit within the 8k limit)
9 different types of policies to guide the agent (json)
3 actual documents outlining domain knowledge (json)
8 Topics (hint harvesting, guide rails, and the pieces of information prepared as adaptive cards for the user)
3 Tools (to allow for connectors)
The whole thing is as robust as I can make it but it still feels like a house of cards and I expect some random hiccup will cause a failure.