Episode 1

Jordan (00:02):
This is a tool we call Ryz Lab's video studio. There's the dog skateboarding in Venice, Italy. Replit is a platform that allows you to vibe code web applications. It's still so early for AI video generation. The more realistic videos look, the more shocking they are.

Jordan (00:27):
Built this week.

Sam (00:36):
Welcome to the inaugural episode of Built This Week. I'm your cohost, Sam Nadler. I’d also like to introduce my cohost and business partner, Jordan Metzner. How are you doing today?

Jordan (00:51):
Hey Sam, welcome to Built This Week. Super excited to kick this thing off. Let's get started.

Sam (00:55):
Let's do it. First, what is Built This Week?

Jordan (01:00):
It’s a demo of tools we've built at Ryz Labs, plus cool AI tools, news, and applications. Every day we’re building, models are changing, and we want to share it all.

Sam (01:29):
These tools are built in under a week. Some are internal, some help grow our businesses, all built with different AI tools. Not simple TikTok hacks — these take a bit more effort but are doable even for non-tech people, right?

Jordan (02:18):
Right. Some technical skills help, but new AI tools let us scale and automate fast. We use these for video, social content, everything.

Jordan (03:14):
Today’s tool made a huge impact on our social teams.

Sam (03:20):
Before diving into the tool, how long did V1 take and what problem were we solving?

Jordan (03:42):
We wanted to avoid expensive subscriptions and give people access to multiple models like ChatGPT, Claude, DeepSeek, etc., through API keys instead.

Jordan (04:15):
This lowered our costs and gave flexibility. The launch of Google’s Veo 3 was a turning point.

Jordan (05:16):
We create tons of content. These tools help us do more, faster, and better — without costly video shoots.

Jordan (06:42):
Veo 3 is amazing but pricey — $6 for 8 seconds. So we built our own generator to give access to many models.

Sam (08:09):
What were the quick iterations, features added after testing?

Jordan (08:35):
We built things like prompt enhancers, image generators, model comparisons, prompt copy buttons, etc.

Sam (10:32):
That “enhance with AI” was key. It improved the quality massively.

Jordan (11:30):
Each model has different speed, cost, and quality. So we compare to decide what's best for the team.

Jordan (13:43):
Eventually, we’ll generate across multiple models at once. One prompt, 25 videos.

Sam (14:47):
What a great tool. What did it take to build?

Jordan (20:00+):
Replit was the main platform — fast setup, built-in backend. Also used Replicate for model APIs. It’s all about integrating fast and testing features quickly.

Sam (28:13):
We used the tool to create campaigns in hours that used to take days or cost thousands.

Jordan (29:37):
Also introduced WhisperFlow — an AI dictation tool. Huge productivity booster, especially for prompts and emails.

Sam (34:02):
I use it daily. It learns custom vocab like business names and helps write faster.

Jordan (35:34):
Alright, let’s jump into the news.

Sam (35:42):
These are fresh stories — let’s go.

Jordan & Sam (36:00–39:00):
Talk about Kalshi raising $100M, OpenAI lawsuits, and Ring’s new AI video features.

Sam (41:22):
Thanks for listening. That was our inaugural episode.

Jordan (41:43):
New episodes every week. Find us on all major platforms. See you next week.

Sam (41:56):
Thanks for listening.

Jordan (41:58):
Thanks.

‍