Another major milestone down this week.

Plus not sure what happened, but had an uptick in projects.

Websites, payments, and automations.

Hopefully GPT can help with all of it.

The topics:

• Payment Screens

• The Next Milestone

• Image to Docs

• Courses Slides

• GPT Pulse

Payment Screens

Back to vibe coding more this week, building out what will be the marketplace of Wave.

I had the backend mapped out to get things started, and the began working on all the screens involved with payments, which is a lot.

Being a marketplace essentially means everything is doubled.

For everything a user could do, there needs to be a screen for the Venue to manage it.

So far the full list looks something like this:

Venue Screens

• Venue Payment management

• Venue Items List

• Venue Item Add/Edit

• Venue Purchases List

• Venue Purchase Item

• Venue Purchase Scanner (scan user’s QR for proof)

User Screens

• Select Purchase Options

• Stripe Payment Screen

• Purchase Success/Failure

• Purchases List

• Purchase Item Details

Even with Stripe handling all the money management and credit card data, there’s still plenty of work on the app side of things.

Venues need to set up a bank account, create and manage what they are selling, like Cover to get in or Tables if they offer that service, and set the price and how many are available, and verify who purchased it.

Users need to be able to buy Cover or a Table, see the price and select the amount, enter their payment information, receive confirmation the transaction succeeded, go to a list of past purchases, and pull up the purchase to prove they purchased it.

Lots of screens, lots of transactions, but the scaffolding is there.

The database is roughed out and the screens are coming together nicely.

I’m sure there will be a number of changes once everything is wired up, but marketplaces require a surprising amount of screens.

Plenty more vibe coding to get Wave properly transacting.

The Next Milestone

Finally reached it!

On Tuesday morning, I was pleasantly surprised to see the small YouTube subscriber count under my channel show “1k”.

Overall the journey wasn’t so bad.

Like anything it just took time and consistency.

On March 10, I published the first video to my channel, and I intended to post at least 3 to 4 days a week.

I quickly realized I had too much going on to maintain that pace so I scaled back to basically once per week, but I stuck with one video per week since then.

Only in the last 5 or 6 weeks did I start live streaming regularly, but as I mentioned last week, all the subscribers came from videos not live streams.

And now, a little over 6 months later, I have 1022 subscribers, from the 3 I had on March 10.

Turns out the general advice is right.

Post regularly and eventually it will happen.

Also helps to be creating useful content in a popular topic.

But I really didn’t spend a crazy amount of time on these videos either.

Most of them were a single session of building the workflow, testing it, and walking through it once it worked.

Still quality and volume knobs to play with.

I could do more content, which I think I will prioritize, but better quality would also help.

So maybe a little of both going forward, but still leaning toward quantity for now.

Luckily seeing 1000 I don’t feel like I’m done, I feel more motivated to keep growing the channel.

So I need to wrap up this newsletter so I can get back to YouTubing.

Image to Docs

I didn’t do a video to celebrate 1000 subscribers but I might sneak one in eventually.

Instead, I recorded a video on how to insert images from Google Drive into Google Docs and Google Slides at the request of a chatter while live streaming.

I was pretty sure I knew how to do that at this point- getting very familiar with Google Drive APIs by now - and thankfully I was right.

It took a few extra steps to get a system that made sense, but I was able to put together an unexpectedly long tutorial to use n8n to update a Doc with a photo.

Kind of crazy how this isn’t more straightforward, or how n8n doesn’t have nodes for a lot of this functionality, but I also get how complicated this gets.

Having to find the index of where to insert the image and also replace text is a bit of a pain.

Google Slides in particular involves a lot of data so it is not the easiest thing in the world, but this system is good enough to get started.

What a way to celebrate 1000 subscribers than to automate some image uploading.

Course Slides

Sticking with the YouTube theme, the live streaming process of building an Online Course Generator system is making decent progress.

Yesterday we finally reached the point where we can have the system generate decent looking, Apple-style slides that can be used as visuals during a short lecture video.

And also quickly hooked up GPT’s tts (text-to-speech) model to create the audio track to go with the course lecture.

Almost all the pieces are there.

The audio track determines the length of the video, so the last part is to figure where in the audio lecture to change slides.

With those timestamps, it will be easy enough to create another n8n workflow to combine the lecture audio with the background images to create a 3 minute lecture video.

And then repeat that process 39 more times.

Here’s yesterday’s live stream to catch up on course automating, as well as live Monday.com file upload troubleshooting:

GPT Pulse

Weird to see Pulse pop up from GPT.

Wave used to be named “Pulse” so I see files and folders named "pulse” all the time on my computer.

But OpenAI’s Pulse launched for pro members this week.

This is something I keep thinking about and overcomplicating in head so I haven’t built it yet.

A proactive AI would be good at this kind of thing - read emails and messages and to-do lists to come up with a daily prioritization of what to focus on.

Having it built into GPT could be really convenient as well because I’m usually chatting with it through the day anyway so it would be automatically updating for the next day if things change based on what I was telling it the day before.

That’s what I would hope anyway.

I saw the notification for this new feature earlier this week but haven’t tried it out yet.

From what I’ve seen just popping it open for a minute, it looks quite useful.

It may finally be the Virtual Assistant that everyone has been claiming AI to be.

The real power is in the daily reminder.

It would be great to have GPT ping me with morning updates and reminders to get the ball rolling for the day, and be contextualized about it, not just something generically inspiring like “hey get after it today!”

Anxious to see if this sticks as part of my daily routine.

May also have to finally use my calendar accurately so GPT can better guess what I’m up to…

And that’s it for this week. Lots of screens for Wave and over 1000 subs on YouTube.

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.

I also have a Free Skool community if you want to join for n8n workflows and more AI talk: https://www.skool.com/learn-automation-ai/about?ref=1c7ba6137dfc45878406f6f6fcf2c316

Let me know what else I missed! I’d love to hear your feedback at @jaypetersdotdev or email [email protected].

Thanks for reading!