Every week I’ll dive into a curated list of links from the latest tech to just interesting reads, as well as updates on my latest entrepreneurial endeavors. Thanks for joining on this journey!

Topics for this week:

• Canva

• The Secret to Modern Marketing

• Local AI Models

• StreamYard

• The General Purpose Robot

Canva

Canva has proven a super useful tool for any kind of “branding” with templates for most popular website profile designs and an extremely easy to use interface. It’s become my go-to free app to create a YouTube logo or banner, Twitter/X banner, or brand basically any kind of profile I use online. This week, I learned they have LinkedIn banner templates as well. I’m not sure why that was surprising, because they have everything, but I was still impressed.

A few weeks ago Canva started advertising “Magic” and I had a feeling I knew what that meant, but I’ve been a little slow to check it out. Their promo video for “Magic Studio” is below, but it highlights how they are rolling more AI into their features. Canva is poised to give Adobe a run for its money with their ease of use combined with these AI features. Not an ad, just generally like the product. Worth checking out if you haven’t yet!

The Secret to Modern Marketing

How about that hook? I’m embracing my inner marketer.

It’s always interesting to me when I find the source of a reference. I’ve mentioned Alex Hormozi a few times if you read my stuff at all, and it’s hard not to run across him if you spend a decent amount of time on the web, but in watching most of his videos, I’ve picked up on quite a few of his references and later discovered where he got those references. In this particular video, he reveals his content strategy and he gives credit to Gary Vee for how to produce massive amounts of content.

The interesting source I found is the original slide deck that Gary released a few years ago showing the step by step process of his content model. Very similar ideas, and it does seem to be a good foundation for a consistent marketing content engine.

It’s something I’m considering implementing myself, only taking a more automated approach. Stay tuned 🤖

Local AI Models

I spent the better part of Sunday hacking away at running an AI (Llama 2) on my computer. It took a while to get a handful of programs to play nicely together but ultimately I was able to get an AI running on my machine, and made it accessible to another app, AutoGen, that would help turn that AI into separate agents that have different tasks, creating a mini AI team, similar to the one I covered last week.

AI models are moving quickly, there’s tons of versions created daily so I ended up trying a handful before I got one running smoothly, but one article in particular lead to my success so I wanted to link that below.

As for the AIs on my machine, I may see if they can take on the task of being a marketing team.

StreamYard

On the continued topic of brand building, streaming can be a powerful tool in the marketing toolbox. Streaming not only lets you interact with an audience in real time, but with some pre-planning, can also produce a video that can be edited down and shared as normal viewing content on YouTube or other platforms.

I setup accounts to start streaming on Twitch and YouTube last week, but haven’t committed to it yet. Streaming on two platforms can be a daunting task, but one trick I think will help manage this effort is that I want to stream on both platforms at the same time.

Something like StreamYard will take your live video feed and rebroadcast it to multiple platforms at once, including Twitch, YouTube, TikTok, Facebook, X, and some others. Plus it will let you manage other aspects of the stream in real time which is super convenient. Although it can handle more than two platforms at a time, I think two is plenty to get started with next week.

The General Purpose Robot

This one stuck with me when I heard it because I hadn’t fully made the connection, but in this keynote from Marc Andreessen, he speaks about how we basically have all the systems needed for general purpose robotics.

AI, like GPT, now has vision capabilities, and can also interpret vast amounts of data, and can understand human speech through voice to text capabilities. Given a mechanical body with cameras, sensors and motors, most of the pieces are there for a robot to walk around and interact with the world.

Below are 2 examples of general purpose robots. One is the robot that Boston Dynamics has been working on for over a decade and the other is the more recent development from Tesla that Marc mentioned. Both bots are already capable of complex, human-like movements and tasks. Paired with a system like GPT, it could be a fully functioning entity.

And that’s it for this week! Keeping the streak alive with 4 newsletters in a row. Those are the links that stuck with me throughout the week and a glimpse into what I personally worked on.

If you want to start a newsletter like this on beehiiv and support me in the process, here’s my referral link: https://www.beehiiv.com/?via=jay-peters. Otherwise, let me know what you think at @jaypetersdotdev or email [email protected] and thanks for reading!