In today's episode, they dive deep into the world of AI, technology, and Ruby on Rails with our special guest, Gustavo Valenzuela. Charles and Valentino unpack everything from community-driven insights and AI advancements to the fascinating board game "Star Realms." They explore Gustavo’s journey, his innovative platform "Startups on Rails," and how it aims to document companies powered by Ruby on Rails. Plus, get ready for discussions on AI's transformative role in development, the economic landscape affecting tech hiring, and valuable entrepreneurial advice for developers. Whether you're curious about Hugging Face's new tools or looking for insights into Rails' vibrant community, this episode is packed with information, anecdotes, and actionable tips you won't want to miss.

Socials

Charles Max Wood [00:00:04]:
Hey. Welcome back to another episode of the Ruby Rogues podcast. This week on our panel, we have Valentino Stoll.

Valentino Stoll [00:00:11]:
Hey now.

Charles Max Wood [00:00:12]:
I'm Charles Max Wood from Top End Devs. And this week, we have a special guest, and that is Gustavo Valenzuela. I hope I got close.

Gustavo Valenzuela [00:00:22]:
Too good. Yes.

Charles Max Wood [00:00:24]:
So yeah. So Gustavo and I had a conversation a little bit ago. I think we met over LinkedIn, I wanna say.

Gustavo Valenzuela [00:00:30]:
Correct.

Charles Max Wood [00:00:31]:
And, yeah, I ran across startup startups on rails. I was like, hey. This looks cool, and I wanted to know what he was doing with it. So we we had a chat about it, and I think you're probably better to tell people what it is than me. So I'll I'll let you take it from here. You wanna just tell people kind of a little bit about your background and then what Startups on Rails is and what you're hoping to do with it?

Gustavo Valenzuela [00:00:53]:
Of course. Thank you, Charles. I appreciate you. First, like I said, thank you for allowing me to be here in your podcast. This is my first one, so I'm excited. Hopefully, the first of of many. Yeah. This is, like, the only type of interaction that we get amongst developers.

Gustavo Valenzuela [00:01:08]:
Right? Developers seems to be a very lonely, lonely profession. So podcast is a a good way

Charles Max Wood [00:01:15]:
to Unless your boss is overly involved in what you're do I won't complain. Never mind.

Gustavo Valenzuela [00:01:21]:
Yeah. Yeah. So, yeah, like you said, I, I'm a, I've been working on startups on Rails for a while. I am, a Ruby on Rails developer.

Charles Max Wood [00:01:31]:
Uh-huh.

Gustavo Valenzuela [00:01:31]:
So I, I guess you can consider me a a new, a newbie or a, how do you call it? The the the new career or, forget the term that they use. But pretty much, I've been, I haven't I'm I'm in the process of trying to get into into into a full time employment. I've had a few, a few contracts here and there, but, you know, I'm trying to get in full time. But I've been doing Ruby on Rails actually since 2013, 2014. That was the first time I learned about Rails, through a boot camp. The first time around, I kind of just wanted to learn so that I can build my own products. That was my original goal. And so I, you know, I did the boot camp and then tried a few different things.

Gustavo Valenzuela [00:02:20]:
Didn't really work out. I got distracted in other areas, so kind of abandoned development up until Mhmm. I think it was the beginning of the pandemic. I don't know if it was before or after. Picked it up again, tried to, you know, get get ready, tried to, kind of learn the the, the holes that I have from the first time. And that went well, and then the pandemic hit us. And there was all these layoffs, so I was, oh my goodness. I came back to development for nothing.

Gustavo Valenzuela [00:02:53]:
But, I started yeah. That's crazy. So, I kept with it, and then I, you know, because there were in a lot of, listings or jobs, I wanted to kinda find out what what the, Ruby and Rails companies in the area. Even if they I figured even if they don't have a a posting, a job posting, I could always, reach out to them and try to, you know, get a connection going. And, I found out that there wasn't any place that I could, see what companies were built on Revlon rails so I can reach out to them. And pretty much only the ones that post, job opening. So I decided to kinda start building, my own database on my own.

Valentino Stoll [00:03:42]:
Uh-huh.

Gustavo Valenzuela [00:03:43]:
And so I figured, you know, I could use that, but I'll this could also be something that other people in the community could use. So I found, the, the u the domain startups on rails.com, and I started to build that as a database of just a documentation of of companies built on Ruby and Rails. And so, you know, I started sharing with the community, decided to get some engagement with the community. I got some feedback, and people liked the project. So I, you know, kept building it. And, actually, I stopped for bid. And, yeah, I stopped for a while. I stopped building it.

Gustavo Valenzuela [00:04:23]:
So now I'm back. I've been building it, like, for the last couple months to to make it into a full on, project that people could use. That's kinda my story more or less.

Valentino Stoll [00:04:36]:
I'm a

Gustavo Valenzuela [00:04:36]:
father of 2 girls, a husband, and, Rubin Rail developer.

Charles Max Wood [00:04:43]:
Awesome. Yeah. Being a girl dad's fun. He did. I think Valentino has kids too, but I don't know if he has girls or not.

Valentino Stoll [00:04:53]:
I've got one.

Charles Max Wood [00:04:54]:
Yeah. I have 3 and 2 boys. So, but yeah. So, way we could talk about girl daddy, but that's that's a whole different thing. So yeah. So first of all, I'm I'm just kinda curious because I love the idea. Right? It's like, who's who's out there using Rails? How do you start figuring out what companies actually use Rails? I mean, some of them are kind of, I guess, well known. Right? There's, you know, Basecamp or Heroku or GitHub.

Charles Max Wood [00:05:33]:
I think Airbnb was started on Rails. And then I also am aware that, like, companies like Twitter, you know, they started out using Rails, but then kinda migrated to, I think, it's a mixed stack. They probably have Rails in there somewhere still because it's hard to get rid of a technology once you adopt it. But I know that a lot of their stuff went to Scala and stuff like that. And so, yeah, how do you decide who's on the list, and how do you figure out who's using Rails?

Gustavo Valenzuela [00:06:00]:
Yeah. So that that was a challenging challenging situation in the beginning. So my first approach was, to kinda manually get as many as I could on my own and then reach out to the community and do some type of outsourcing or crowdsourcing and get people to kind of, submit their, either companies that they work for or their own projects or composite companies they know of. And that worked to a certain degree. But, you know, I got a couple 100 companies that way.

Charles Max Wood [00:06:34]:
Right.

Gustavo Valenzuela [00:06:35]:
So what I'm doing now is I'm trying to do some, scraping and looking through different websites and listings and things like that. Ultimately, I think there's about 700,000, website built on Ruby on Rails, I believe. So that's gonna be a challenging task. So I I'll take any any ideas or input, from the community.

Charles Max Wood [00:07:01]:
Is there a way for people to submit? Hey. We use rails. Right?

Gustavo Valenzuela [00:07:04]:
Yes. Yes. Yes. Yes. Usually, what I do right now is I periodically, I, you know, post a tweet, or x and ask people to to give me some ideas. I should have built something where people can submit on my on the website, but I haven't built that yet.

Charles Max Wood [00:07:27]:
Yep. But,

Gustavo Valenzuela [00:07:28]:
so far, it's just really through Twitter.

Charles Max Wood [00:07:32]:
They use rails at proximity, don't they, Valentino?

Valentino Stoll [00:07:36]:
Yeah. We do. Pretty heavily.

Gustavo Valenzuela [00:07:41]:
Medicine. Okay. Yeah. I noticed that there's a a number of medical companies or startups that use rails. Uh-huh. At least a handful I've seen of pretty big, startups.

Valentino Stoll [00:07:58]:
This reminds me of, so I don't know if you know, but Andy Kroll has recently, like, published, using rails.com. Yes. I do like kind of how you how the layout of the of your site works, though. What are you, like, pulling information from that data source as well? Or

Gustavo Valenzuela [00:08:25]:
I'm pulling in for information from a lot of different sources. Yeah. And I do I mean, he's one of the, 10 that just recently popped up, which kind of

Valentino Stoll [00:08:39]:
Right. I know. There's so many now. Yeah. I started

Gustavo Valenzuela [00:08:42]:
to like nothing.

Valentino Stoll [00:08:46]:
I guess I where, like, where does this, like, I guess I'm try I like, I the design I use is awesome. Mhmm. I wish I had seen this, to be honest, like, so many times in my career. Right? Being able to to filter it by different things and tooling and things like that. Like, how how do you, like, how do you think about, like, the consolidation of these different companies and, you know, maybe, what are your plans for it, I guess?

Gustavo Valenzuela [00:09:30]:
Yeah. That's a good question. So I I just recently noticed that there's a number of companies. So I kind of like I don't wanna build something that's already you know, doesn't have a utility or, you know, a lot of people are doing the same thing. So I think there's opportunity to build on top of this database. Right? To to build, features that could be useful for a community. Like, I mean, the lowest hanging tree is just categories. Right?

Charles Max Wood [00:10:01]:
Mhmm.

Gustavo Valenzuela [00:10:01]:
But, also, I want to know I don't know. Even categories that are more, outside of the typical categories, like, I'm thinking companies that are more active hiring junior developers, you know, or companies that are more active, supporting, open source. That type of categorizing. I think, just given the, the users kind of options as to who they would wanna support. Right? If in or or or another category would could be, you know, the substitutes for the rail substitutes or rude substitute for such and such, products. But, there's, I think, opportunity to to create content on top of that. Content, targeted towards, not necessarily developers, but product product people or or business people. Just marketing the tools and the framework in general.

Gustavo Valenzuela [00:11:06]:
That's just one, but I've you know, I have a list of ideas that I could build that I could build on top of the the, the bay the database. Yeah. Another thing that I one of the things that I want that I'd like to do is kind of get involved with the developers in a company. So one of the features that I'm working on right now is, adding the ability to for companies to list open source that they're working on. So if and if they need help with whatever open source. So but through that, if I'm a developer looking to connect with other company, I can go to the company, see if they if they have some open source open source issues. I can, you know, start participating through that and maybe through that, hopefully, make a connection with that, with the company, if that makes any sense. I don't know.

Gustavo Valenzuela [00:12:12]:
There's a lot of things that I I feel like could be built on top of just a pure database.

Valentino Stoll [00:12:19]:
Yeah. You know, I love seeing that, on the company view. It would be really cool if you could, like, submit, like, a, you know, a pull request right from the page. Mhmm. And say, hey. I'm, I'm up to I'd like to update the information here. Like, have a sign in with GitHub kind of thing so you can verify that they're, like, part of that organization even. Right? Because I mean, like, it it definitely, like, feels like that's, like, what you could do.

Valentino Stoll [00:12:47]:
Like, Yeah. I don't I don't know. Like, how do that's the hard part with, like, all of this stuff is, like, you know, it's just you. Right? Mhmm. Like, community, like, you know, how do you get, like, the community involved in all of these different sites that exist and garner support? Right? Like, where where do you find, like, the most, like, people want to, like, help contribute to so far?

Gustavo Valenzuela [00:13:20]:
You mean in my personal project or just in general?

Valentino Stoll [00:13:24]:
The 1st startup on rails.

Gustavo Valenzuela [00:13:25]:
Yeah. Yeah. Yeah. You know, we're so I haven't really to be honest, I the last couple few couple months, I haven't been interacting, but that's some it's a good question I can ask the community as to where, direction they would like for this to go. But, and then you're asking in terms of, like, having people, contribute to this project, or is that the question, or or just what people would like in terms of features?

Valentino Stoll [00:14:02]:
I I guess I I'm interested in both, but, I was more, like, where people are are currently, like, interested in contributing to ad, stuff. Like, do you see do you see people are, like you know, they reach out to you about wanting to update specific things, or, is it pretty low volume at this point?

Gustavo Valenzuela [00:14:23]:
Yeah. To be honest, I haven't, yeah, I haven't really, launched officially, so I haven't put it out, of there. But that is a good question I haven't asked. I haven't seen any I haven't had had any request. And, also, the idea of maybe open sourcing this project sounds interesting, right, so people can,

Charles Max Wood [00:14:49]:
you

Gustavo Valenzuela [00:14:49]:
know, contribute directly. That would be a good idea. I would have to clean my database. It's a little embarrassing right now.

Charles Max Wood [00:15:00]:
That's that's a story

Valentino Stoll [00:15:01]:
of,

Charles Max Wood [00:15:01]:
like, more than half my code. It's like, oh, there's a ton of junk in here.

Gustavo Valenzuela [00:15:06]:
Yeah. Yeah. So that's that that that but that's an interesting idea, and it's another, another way to gain experience, right, in managing this open source project would be interesting. Yeah.

Charles Max Wood [00:15:22]:
Yeah. I'm I'm also curious. So you mentioned that, you wanted to know what companies were kinda local to you that were using Rails, you know, maybe for a job search or something. But, you know and that kind of appeals to me. I I'm curious also just from kind of the broader view because a lot of people that I talk to are, how do I put it there? They're kinda looking for their next position or their next, you know, whatever, their next job. And, yeah, it's just you know, I think something like that would really help. But, also, with all the layoffs and everything else, do you see the rails market shrinking or growing, or do you have a a feeling about that?

Gustavo Valenzuela [00:16:17]:
Yeah. I mean, I just know that, about a year about 18 months ago, I have recruiters hitting me up. You know? I wouldn't normally make it through the whole interview, but right now, you know, I'm trying to apply and do that. It's it's hard. I mean, in terms of the feeling, it doesn't feel like, there's openings. You know?

Charles Max Wood [00:16:40]:
Yeah.

Gustavo Valenzuela [00:16:42]:
Yeah. So I don't know if, it's a good time. If the openings are there and you just have to directly kind of get in touch with the company and and see if you can get in that way, or they just flat out no openings or no opportunities. But in any case, it's always better to to try out different strategies, right, on your

Charles Max Wood [00:17:04]:
Yeah. Well and it's it's also interesting because I think there are companies that use rails that could use the help, but because of the way the economy is sitting right now, it's, you know, it's it's just hard for them to be able to afford or justify the additional headcount. Right? It's not, hey. There's no work to be done. It's just, you know, given the current state of things, we're not comfortable, you know, making the commitment to pay another salary given our you know, where where we see finances going and where we see the, the economy going long term because, yeah, if you wanna stay open, you kinda have to plan ahead with your cash flow, and it it it's hard.

Gustavo Valenzuela [00:17:54]:
Yeah. Yeah. It's, it's just a matter of numbers. Right? It's, math. Yeah. Yeah. I think to me that, you know, that the economy together with AI, which I've been using, you know, the last few months Mhmm. More and more, I do think it makes developers way more effective.

Gustavo Valenzuela [00:18:13]:
Right? Way more Yeah. I know for me, it saves me a feature that I could do sometimes it'll take you 2 days to do. I can do it in half a day. You know? And that's Mhmm. Me being, like, super careful, you know, looking at everything and trying to analyze everything and not really knowing my way around. But I think that it's a fact to me that some that, you know, needed, 2 or 3 developers. Maybe not one developer can do it, and that's just a reality. But, Yeah.

Gustavo Valenzuela [00:18:49]:
I think a a a good way I for developers, like, that are unemployed, I think a good alternative is building your own products. Right? Building something in this. Right? You know, become entrepreneurial.

Charles Max Wood [00:19:05]:
Yeah. I agree.

Gustavo Valenzuela [00:19:08]:
Yeah.

Valentino Stoll [00:19:12]:
So how did you build this thing? What like, what's, what's the stack? What did you go with?

Gustavo Valenzuela [00:19:16]:
So Ruby and Rails, that's what I've been doing for a for a while. It's the fastest thing. You know?

Valentino Stoll [00:19:22]:
Are you using all the the shiny new tools, like the all the solid stuff?

Gustavo Valenzuela [00:19:28]:
No. I have I haven't even gotten into into that. Wow. Yeah. Yeah. I feel like there's so much. There's so much. It's

Charles Max Wood [00:19:34]:
got a hard

Gustavo Valenzuela [00:19:36]:
with everything.

Valentino Stoll [00:19:38]:
I feel like that's one thing that's missing that maybe some magic contributor out there will contribute. But, like, just like all the features of Rails, like, pick them apart, look at all the lists, like, you know, editors, like, you know, drop down selects, forms, like, all the different things that you could do with it just like that you can just see better. Because I feel like the guides are a great start, but, like, it's more granular that than I want. Right? Like, I want okay. Like, I'm working on email stuff. Like, what features do I have available in Rails? Right? And then I feel like there's so many just for email. Right? Like, it would be interesting to see a visualization of that.

Charles Max Wood [00:20:25]:
So you you're talking about, like, features in Action Mailer, and then there's also the other one that reads the emails.

Valentino Stoll [00:20:35]:
Yeah. And even on a a, you know, a a more scoped level of, like, okay. Like, I have, like, a, you know, an active record model. Like, what are all the features that I can inject into this model? Right? Because I'm using active record as, like, a base class. Right? And, like, there's so many, like, DSL methods you could use that are specific to that. That would be nice to just like, I know, like, the guides, like, have a good, like, general overview of, like, associations and, like, smaller pieces. Right? But, like, I don't know. Like, what about the more specific, you know? I don't know.

Gustavo Valenzuela [00:21:16]:
Yeah. Yeah. Different use cases. You know, it's I think it's like Different use cases. Yeah. Is is it double h so you know, is is is the that way it's sort of with frameworks like rails. Right? But it's so good for you because it gives you everything, but but in the other hand, it's too you know, it gives you everything. So there's tons and tons and tons of, features and, aspect to it that are you know, it's hard to know everything.

Gustavo Valenzuela [00:21:43]:
But I think that almost gets solves gets solved by, AI. Right? AI is good at, scrolling through everything, and then we just verify that it's indeed what we need. You know? I think that's AI and rails is it's it's a good pairing.

Charles Max Wood [00:22:04]:
I I would I I would say that that is kind of true. It depends on what they're training it on. Right? So, like, Action Mailbox was the other library I was trying to think of, right, where it effectively, it goes and it grabs the emails, and then it essentially routes it through something that's sorta like a controller. Right? And so then it does things with the emails, which I think is really slick. Right? And that's that's something that I can think of a couple of use cases for. But, you know, I don't know if a lot of people are using it. And so I don't know if the code that they have fed into the AI large language models to say, hey. This is a good, you know, a a good example of how to do this.

Charles Max Wood [00:22:59]:
Right? Because it has to train it a bunch of times. And so if you ask it for our action mailbox code, it may not be great. Right? But if you're looking at some of the more common stuff like active record and a lot of the more common use cases that people put active record to, I would imagine that that's probably a lot more commonly gonna come up and and be handled well by, you know, by act by, the large language models. Right? Because you're gonna you're gonna have a zillion examples because everybody's using models. But even then, like, if there's some obscure feature that not a lot of people use, even if it's a powerful one and could do good things for your app, it it may not opt for it because it's been trained over and over and over again on how people actually solve the problem. And so, yeah, I I can see where it comes in and is helpful, but I think I think, yeah, the the less well used features, your AI may not be trained well enough to pick it up and give you the option.

Gustavo Valenzuela [00:24:10]:
Yeah. No. I agree. I think you still need to be, you know, knowledgeable.

Charles Max Wood [00:24:14]:
Yeah.

Gustavo Valenzuela [00:24:15]:
But, I think it's hard to keep all that overhead in mind when you're coding. Right? There's so much going on. So that's a good way to kind of, like I mean, AI does well at thinking of everything and shrinking it down, and then we Yeah. Better at verifying. Right? We've been verifying.

Charles Max Wood [00:24:34]:
Well yeah. And and I'm not saying that you're wrong to do it however the AI tells you to if it works because it does save you time and effort. Right? You don't have to go look up the API and make sure you have all the arguments in the right order and, you know, whatever else. And there there are handful of, common things in Rails. Like, I can never remember if it's options for select from collection or options from collection for select, and I always screw it up because I always guess it wrong. Right? And so having the AI go, oh, I know what you're doing. You know, that's that's kinda nice. But yeah, you know.

Charles Max Wood [00:25:13]:
And so then yeah. Sometimes it'll give me a longer snippet of code, and I'll wind up tweaking it because it's like, yeah. This is close, but it's not exact. Yeah. And Exactly. And so yeah. So then it saves me a bunch of time. And even if there's a more elegant solution, it doesn't matter that I don't know it because I have something that works.

Charles Max Wood [00:25:31]:
I can look at it and visually parse it, and I didn't have to spend 2 hours figuring out how to do it the other way.

Valentino Stoll [00:25:38]:
Yeah. Speaking of of AI, it it makes me, think about your your new start up on Real Site where you have, like, a Real's AI category. And it makes me, like, imagine like, I I would love to be able to see, like, what companies are using what has to so rails. Like, not only just, like, rails and AI specifically, but, like, alright. Is anybody actually using Action Mailbox in a production app? Like, how are they using it? Like, are they willing to disclose that, like, just in an itemized list? Right? Like, that would be super helpful, like, not only for, like, the rails community in general. Right? Like but, you know, how how are people building stuff? Right? Like, exposing, like, the use cases because, like, there's so many. Like right? Like, it seems like there there are so many that, like, also give talks at these conferences all the time, and you think, like, there it would there would be, you know, a way to expose that other than just, like, publishing a a conference talk. You know? Yeah.

Valentino Stoll [00:26:46]:
You know?

Gustavo Valenzuela [00:26:48]:
1000%. Yeah. That's exactly why. I kinda have you there in the corner because I haven't gone in that direction, but, you know, that's a completely different thing, you know, doing a little research and, but, yeah, that's that's something I'm interested to. And I'm figuring out who's using it and how they're using it. Figuring out the companies, but also like you say, the details as to, how they're using it. Yeah. That's something that I'll definitely work on.

Gustavo Valenzuela [00:27:15]:
And, I mean, that's the newest thing. I could probably be working on it right now. That's a good opportunity for content.

Valentino Stoll [00:27:21]:
I mean, you you can start, like, you know, ingesting the conference video talks and just, like, have AI, you know, categorize which company they work for and what they're working on.

Charles Max Wood [00:27:36]:
Yeah. Yeah. One one other thing that comes to mind that would be interesting with this and I know we're kind of brainstorming this, but, I mean, you know, you said your prelaunch and, you know, these are all ideas. And maybe we could talk about what the minimum viable product looks like, but, I keep hearing from people that Rails is dying. And, you know, I I understand some of the arguments that people make, but I still see a lot of companies starting up on Rails. And I also see a lot of companies that have adopted Rails or that started on Rails, you know, 5, 10, 15, 20 years ago, that are doubling and tripling down on it. And so I I don't know that and maybe maybe we just need a better definition of what people mean by dying. But it'd be interesting to see, okay, if you start a start up with Rails, right, then, you know, then it gets listed.

Charles Max Wood [00:28:38]:
In that way, we can see. And then some of the other companies, right, they come in and they say, we started in, you know, 2020 or 2021 or 2022. Right? And so you can see how long they've been out there and, you know, running things on rails. And so that way, we kinda get a feel for, okay. Yeah. There were, you know, a gazillion companies that started up in 2,008, 2,009, 2010. Right? And then you get into 2016, and maybe they're they're, you know, half that many. Right? You get to 2020, and you can see, oh, there's an uptick here for whatever reason.

Charles Max Wood [00:29:14]:
Or, it it'd be really fascinating just to kinda see how that all comes out. And then people can see, oh, okay. We've got you know, we we have a list of, you know, a 100 startups that started in 2024. Right? And more getting added all the time. And, you know, of of the 120 of them are venture found funded. Right? Which means that they I I I was gonna say they have a better chance of making it, but that's not necessarily true. But, you know, so many are venture funded. So many are, kind of micro SaaS, kind of tiny services.

Charles Max Wood [00:29:56]:
So many of them are right? And so then you can get an idea of, oh, okay. So this is where this is where the action's at, and we can see what the trend is. Right? We can say, oh, okay. We're we're getting more companies in in, you know, also in industries, right, where which you're already doing. Right? We're we're seeing more health care, apps starting up, or we're seeing more education apps starting up, or, you know hey. Given the current political climate, we're seeing a lot more political bent apps coming out or, you know, whatever. So, anyway, I think that'd be really, really interesting just to kinda see what that trend line looks like because it's it's kinda hard to know. But if you give people a way to self report

Gustavo Valenzuela [00:30:47]:
anyway. Yeah. Yeah. No. I I agree with that. I in the data aspect of it is some that is, I'm fascinated by. You know? And the idea that we can it can them way more granular granular that how we, you know, we we, process or or play with the data, you know, in terms of, like, when they started, how are they doing? You know, what type of volume in terms of sales or revenue and things like that? Specific industries and things like that. To me, I find that very interesting.

Gustavo Valenzuela [00:31:20]:
And I also think that it could also influence, potential entrepreneurs, right, that are trying to build their own products. We can, kind of tailor that data and into content and put it out there through you know, and see if people can make decisions based on the the data.

Valentino Stoll [00:31:38]:
Yeah.

Charles Max Wood [00:31:41]:
Yep. Yeah. I think a lot of it just comes down to what questions are people going to ask of a system like this, and then what data do you need in order to answer them.

Gustavo Valenzuela [00:31:54]:
Yeah. Yeah. It's a chicken and a chicken and a neck problem situation. Right? That's a

Charles Max Wood [00:32:02]:
Yeah. I I think if the community just fully adopted, you know, startups on rails or, you know, something like it and, you know, everybody just kind of commonly submitted to it and kept it up to date, I I think it would be very interesting just because I think we would be getting decent data on that. And I don't know that every company is gonna be willing to tell you, hey. We're using these gems or these features of Rails. But, you know, the ones that are willing to share it, you can at least give us some idea what's going on. But as for, hey. We've been around this long, and we're using Rails. Right? Or maybe we've been around this long, and we've been using Rails, and we had a a major rewrite or a significant chunk of our stuff was kinda reworked at this point.

Charles Max Wood [00:32:56]:
Right?

Valentino Stoll [00:32:57]:
Mhmm.

Charles Max Wood [00:32:58]:
That might be interesting. But, again, you know, even if they just put in, we started in 2,006, We've been using Rails the whole time. You know? We're a company that's x so many big. Or maybe you go ask LinkedIn, how how big is this company? Then, right, then you get an idea of of what we're looking at.

Gustavo Valenzuela [00:33:19]:
Yeah. That that is definitely interesting. And I think what I found over time was that, you need more compelling features or reasons for people to come back to the to the website, both the users, right, and the companies. So and that's kinda why I get I get trapped in. Okay. What can I do to make it compelling, to make it more appealing for people to to wanna participate, both in companies and end users? So one of the things that I was thinking of, you know, is perhaps using the company page as a quote, unquote engineering page that a lot of websites have. Right? Like, you go to Airbnb slash engineering, you'll see blogs, related to engineering and you see like, when you just talked about, like, more detailed things about what they've been doing. They they post their open source, you know, stuff that is related to, to the engineering community.

Gustavo Valenzuela [00:34:26]:
And, I think that might be a service that could be offered to smaller companies. Like, if you have the Airbnb's have the resources to have these type of pages, but this could be something that targets, smaller companies that, you know, a place where they can post, data or content related engineering to engineering. So I don't know. Yeah. The challenging has been part has been that. Like, how can we, compel companies to to participate and, through that also get users to wanna visit. So she's gonna make problem as I said before.

Charles Max Wood [00:35:13]:
Yep.

Valentino Stoll [00:35:15]:
I mean, to your point, though, like, a lot of companies, they want, right, like, exposure. They want other developers to know what they're working on, what they're doing in the community. Right? Like, they're despite what many people may say about, like, the, you know, hiring market for for any programming, but, like, Real specifically, right, like, they still want that exposure. Like, there's gonna be a point where, you know, hiring is hot again. Right? Like, it's it's it goes in cycles. You know? Like, there are always lows, and there's always highs. Like, and, you know, as, like, any company that is, like, you know, tight on resources, which, you know, from, like, a developer standpoint, right, like, which just happens. Like, eventually, like, once you're to a certain size, you're, like, gonna have this point where, okay, you've been developing this new thing for so long, and then it releases.

Valentino Stoll [00:36:19]:
And you're like, oh, crap. Like, everybody's using it, and we need more people to manage this thing that we built. Right? Like, I feel like it's just like a thing that happens. And and you want as a company to, like, get the exposure to the people that, you know, could potentially be working for, you know, the company. I don't know. I feel like as a developer who has looked for a job before, right, like, I've I've looked to see who's, like, you know, doing the conferences, who's, like, present in the community, right, like, who's open sourcing stuff. Like, these are all things that I think about personally. And it's all just a matter of exposure for the company.

Valentino Stoll [00:37:00]:
Right? And there's honestly a lot of great companies out there that don't do that. Right? And is there a way where that could process could just be a little easier? Right? I think it's what you're getting at. And I think that's completely the case. Like, making it just like a a uniform way where people can, like, you know, publish numbers to some, you know, public public place, I feel like, is what what's needed, really. We we're still lacking that in a lot of ways. Yeah.

Gustavo Valenzuela [00:37:36]:
Yeah. So yeah. That's a that's a that's a direction. Right? But to me, I'm the type of personality that I want everything. So I I suffer with paralysis of analysis stuff. So, oh, that that's a that's a good feature, and I end up wanting to build everything and then end up build no. Building nothing. So but, I think that's a good strategy for, like, an MVP, you know, going in that direction, allowing companies to to build their own profile and post their technical, content.

Gustavo Valenzuela [00:38:13]:
That's something that all services need to think about.

Valentino Stoll [00:38:18]:
Yep.

Charles Max Wood [00:38:21]:
So I'm I'm curious as we're kind of digging through this. I think we've had a lot of ideas around what we could do or what you could do with start ups for on rails. But I'm I'm a little curious, like, what other resources that are kinda like this, would you like to see for the Ruby or Rails community? Because it it seems like there are definitely opportunities out there for us to provide something that allows Rails developers to get information on or collaborate on or, you know, whatever on other aspects of the community?

Gustavo Valenzuela [00:39:04]:
Yeah. So I think one that comes to mind right now, and it's not necessarily related to these, it's more on the the way of helping, new commerce or new developers.

Valentino Stoll [00:39:21]:
Mhmm.

Gustavo Valenzuela [00:39:26]:
Navigate through what I call the last mile of developing, the last mile of learning. Right? That last mile where it's like, okay. You've you've learned. You've been through all these boot camps or self learning. That last piece of learning that is gonna give you the ability to be ready job ready. Right? Which I I think that content is hard to find. One of the things that, I was part of was the agency of learning. A friend of mine, Dave Paola, he started, which is just pretty much, get a group of people together, a group of young developers and a few senior developers.

Gustavo Valenzuela [00:40:10]:
And what we did was, participate, in open source. We helped, you know, contributed to open source, but we did it in a way that was more structured in that, we we we almost did it like a job. Right? So, pick a issue, you know, submit a, a proposal, and the proposal is gonna be not necessarily code, just like, what are you what's the strategy on how you're gonna solve this issue? There's a code review, the community code review and you know, where they pick you know, go at you and ask you questions. And whilst you have a pretty solid proposal, you're going to, you know, submit the you know, work on the issue and and then submit and get a review. So I think that was very helpful in terms of getting, like, almost almost professional experience. Right? I don't know if I I haven't I haven't really worked for a company, but I don't know if that's the way it would work. But it felt like, okay. This is it feels like I can talk about, something that looks like a professional experience.

Gustavo Valenzuela [00:41:18]:
So that was a way that, they've kind of visualize this issue or this problem, the last mile. But, you know, I think there's something there, like, helping people navigate through the last mile. But I also know that the, you know, the the the the state of the industry is not necessarily favorable for a new developer. So so I don't know if that's a it's good timing for that.

Valentino Stoll [00:41:51]:
Mhmm.

Charles Max Wood [00:41:53]:
I like the idea of kind of the last mile. Right? It's okay. You can build a Rails application, but here's the rest of the stuff you have to be able to do or know how to right? That they don't necessarily teach you or spend a lot of time on.

Gustavo Valenzuela [00:42:08]:
Correct. Yeah. And a lot of these are more intangible, you know, things that you gotta wrestle you know, you gotta wrestle with, you know, the dynamics on on your, amount of your, you know, your engineer, amount of your Mhmm. A lot of you, this is emotional. You get some feedback. And, like, dude, I worked so hard on this. What do you mean I still need to do more? And it has to do with my preferences. Like, I prefer it this way.

Gustavo Valenzuela [00:42:31]:
You know? I know it works, but this is the way I prefer. Things like that that are not very tangible. That was helpful to me.

Charles Max Wood [00:42:39]:
Right. Very cool. Well, I told you guys I have a hard stop coming up, so, I'm gonna push this over to Pix. This has been fun to talk about. I mean, one other thing that I've been looking at putting together actually is a directory of learning resources for Ruby and Rails. Right? And so, essentially, it would list. And and I've I've started on the code base, and I've been I've been working on it for a while. I just haven't launched it yet.

Charles Max Wood [00:43:05]:
But, and I've based it on our picks, so it's called rubypicks.com. But the idea is is yeah. So it's here, you know, you can search podcasts or podcast episodes. You can search, YouTube videos, conference talks, you know, blog posts, documentation. Right? And it'll you tell it kinda what you're looking for, gems, so RubyGems. And then it'll list out all the resources out there, and then people can rate and review them and things like that so that, the ones that get more highly rated you know, and we also try and attach versions, right, so that, you know, you can put in what version of Rails or device or whatever you're using. But, yeah, then then it'll come back and then say, okay. These are some highly rated resources that align with what you're doing.

Charles Max Wood [00:44:01]:
And so then, you know, maybe you get the walk through on how to set up Stripe or the walk through on how to whatever. And the other thing is is that, I would like to share some of the search data on the back end that says, essentially, these are the most common searches that don't have a lot of results. And that way people know, oh, I could do a video on this. Right? Or, hey. I could write a blog post on this and just kinda spur the community into taking action and providing that kind of stuff. So Oh,

Gustavo Valenzuela [00:44:37]:
sounds like a good idea. Yeah. I like that. So people don't waste their time and yeah. It happened to me too that you wasted time working on on material that might not be as relevant. But if you know what's number 1, number 2, number 3, you know, that's a good idea. Yeah.

Charles Max Wood [00:44:53]:
Yeah. Yep. Anyway, let's go ahead and do picks. Valentino, you wanna start us off?

Valentino Stoll [00:45:01]:
Yeah. Sure. So all my picks are AI related today, probably as usual. But I've been, the folks at Hugging Face are just, like, on fire lately, and I've been experimenting

Charles Max Wood [00:45:18]:
much stuff there.

Valentino Stoll [00:45:19]:
So much stuff. I mean, they're just announcing stuff like crazy. Yeah. And I've been playing with, they have transformers dot JS, which basically lets you run inference using the Onyx runtime of LLM models in the browser, using WASM. And so you can download various models from Hugging Face and perform the inference of various features, like summarizing or categorizing stuff or all kinds of different, pipelines that they provide. And you just do it all in the browser. So if you have sensitive information, you don't want, like, pass to servers, and you want to do some stuff in the browser, it it works, really well. I even got it to do a rag, in the browser.

Valentino Stoll [00:46:08]:
Just Oh, wow. Fully local in the browser doing embeddings, and then searching it and, running a small model, to to do the generation. But, it it works, and it's just like crazy.

Gustavo Valenzuela [00:46:22]:
Wow.

Valentino Stoll [00:46:23]:
So, I'm excited to mess around with that more. So that's transformers JS. They have a GitHub for it and everything. And the other one that they, they also released, I'm trying to get the name of it, but they have a so they have a leaderboards that they manage where they run all the large language through different arenas and rankings. And so they've open sourced the pipeline that they use for running those evaluations and, have open source kind of to try and garner support from the community to, like, publish different open, valuation datasets and, algorithms and and things like that. And it's just really awesome and well done. I think it's called, light eval. And then on top of that, on on the evaluation side where if you're not familiar with evaluations, it's just a matter of, like, testing whether or not the outputs do a specific thing, and you can evaluate them in a lot of different criteria.

Valentino Stoll [00:47:36]:
But they also published a a a guidebook, which has, like, a huge, just data bank of different ways that you can evaluate, LLMs. And so LLMs as a judge, human evaluations. They have benchmarking, and it's just like a great, like, go to guide for, doing all of this stuff.

Charles Max Wood [00:48:01]:
Interesting. Awesome.

Gustavo Valenzuela [00:48:06]:
May I ask a question, regarding your, the first, one that you picked? What kind of machine are you using to run, that, model? Are you running it from your local machine? Or

Valentino Stoll [00:48:16]:
It's local machine. I have an m one MacBook. I have not tried it on a lesser machine. So I I am curious how how well it performs on maybe, you know, a Chromebook or something like that. I haven't tested it.

Gustavo Valenzuela [00:48:32]:
Yeah. M one. And and then what kind of are, RAM do you do you have a good RAM for that? Because I think it's it's is RAM a pretty important, resource for that?

Valentino Stoll [00:48:46]:
So they, they do a combination blend of GPU and CPU.

Gustavo Valenzuela [00:48:52]:
Okay.

Valentino Stoll [00:48:53]:
And so it's primarily, like, the GPU and CPU that, get the most hit out of it. Browsers do let you kind of, like, download almost infinite amount of, content to the browser's cache, which is nice, but, you know, so it's more of like the the storage is not necessarily an issue because it'll store it, just in the browser, as, like, a staged aspect for the for the model itself. But, yeah, that's a good question about memory. I don't know I haven't done any benchmarks yet on, on the memory consumption.

Gustavo Valenzuela [00:49:32]:
Wow. Thank you.

Charles Max Wood [00:49:36]:
Cool. Yeah. I'm I'm diving in. Are are you done? Because I can just roll into my picks otherwise.

Valentino Stoll [00:49:46]:
Oh, yeah. Yeah. That was it.

Charles Max Wood [00:49:48]:
So I I usually do a board game kick pick first, so I'll just throw one out there real quick. This is one that I've actually played on my phone and my computer and not with actual physical pieces. It's Star Realms. I don't know if you all played that, but it's a card game. And it's got a whole bunch of different, sets that go into it, which makes it fun. But, anyway, you can pick it up. You can play against people on the Internet. And, yeah, it's a deck building game.

Charles Max Wood [00:50:20]:
So you, you know, you draft cards and you play cards and you, you know, you try and get good combinations of the cards that work together. Usually, it's by playing multiple cards the same color, but not always. So, anyway, I'm gonna pick Star Realms. I didn't look it up on BoardGameGeek, but I I know that you can get physical cards. And, I'm sure that they have it on here. And just let me look it up. Yeah. They've got a 20 a 2014, which is the the original version, and then you also they have a 2021 version that has, extra player decks and the play mat and stuff like that.

Charles Max Wood [00:51:18]:
And then there are a whole bunch of expansions. There's Gambit, Colony Wars, Frontiers. Anyway, the base game is rated at, weighted at 1.92. So, you know, casual gamer, it's, more complicated than your your kinda really, really, really simple games. But it's it's definitely learnable for most people who just want a casual game. So, I I tell people that 2 is about where, you know, kind of the the casual game with enough complexity to make it interesting, but not so much that you have to actually go back to the rule book periodically to make sure that you're doing it right. Anyway so so yeah. So Star Realms.

Charles Max Wood [00:52:07]:
And then, yeah. So I'm gonna do kind of the shameless self promotion stuff. I'm also trying to find a website for Star Realms, because I know that you can just play it online. You can get it on Steam, I think, and then there's the mobile app as well. Anyway, so I've been talking to a lot of folks. I haven't gone as deep as, Valentino has with the AI stuff, but, I've been I've been getting into it a bunch. And I I recently had my contract end, and so I have a lot more time to start really diving in and building out what I wanna build. But a lot of folks are interested in learning it.

Charles Max Wood [00:53:00]:
And so what I'm working on right now is actually building out a a handful of, AI systems, mostly kind of the agentic AI sort of things that we talked to Obi about. And I'm gonna be putting on a boot camp, at the beginning of next year and just walking people through a lot of this stuff, you know, a lot of the approaches, to setting up these agents and then, you know, kind of fine tuning what they they get. So keep an eye out for that. I do have a domain that I purchased for it, but I haven't set it up yet. So, I'm probably gonna just start dropping ads into the shows, you know, in lieu of some of the ads that wind up getting put in that I get complaints about. So, just keep an eye out for that. I think it was, agenticaibootcampdot com or something like that. But, anyway so I'm doing that.

Charles Max Wood [00:53:59]:
And then, yeah, I'm I'm kind of overhauling, top end devs and, you know, updating the member's area and things like that. And so, you know, you'll you'll have a a new experience for the inside of the boot camp. The way that I'm gonna run it, though, is between now and Black Friday, you'll you can get the Black Friday deal. So I'm looking at, running, essentially, weekly help sessions. Right. So you'll you'll get the curriculum for the boot camp every week, and I'm looking at 3 or 4 months. We'll have weekly calls for 6 months, and, I've the price I'm looking at is $35100. But if you do Black Friday, you get $2,000.

Charles Max Wood [00:54:54]:
I also intend to do an AI summit, in December or January, and you will get a ticket to that if you sign up for the Black Friday. So you get you get the discount, you get the AI summit, and then you you get like I said, you get the calls for 6 months. And so, yeah, that's that's the plan there. Did I say that the discount was for $2,000 for the Black Friday? Anyway so that's that's that's what I'm working on now. I have to say that I also just, did a Stripe connect. I had a Stripe connect to top end devs so that I can because there are other, podcasts that I wanna start that are not programming podcasts. And so I don't wanna put them on top end devs because then it's like, oh, there's a political podcast on here. Hey.

Charles Max Wood [00:55:53]:
The you know, Chuck's Church stuff is on is on here. Right? So, yeah, so I'm looking to spin some of that off. And so I was like, okay. I wanna do multi tenancy, but then I, you know, I want them to kind of run independently on the financial end of things for whether it's sponsorships or people buying courses or anything like that. So, anyway, Stripe Connect is pretty cool, and it's not I was worried it was gonna be super complicated to set up, and it wasn't. So, anyway, pretty happy with that. So I'm gonna I'm gonna pick Stripe Connect as well. Alright.

Charles Max Wood [00:56:31]:
Gustavo, what are your picks?

Gustavo Valenzuela [00:56:33]:
Thank you. So for me, in terms of, you know, outside of development, it's just children. Right? So, I'm gonna pick, Disney Princesses. Disney Princesses is my, pick of the week. I, spent the last 3 days in, Disneyland with, about Disneyland Hotel.

Charles Max Wood [00:56:57]:
Oh, fun.

Gustavo Valenzuela [00:56:58]:
Yeah. It was good. It was good. And, we, took our 2 daughters to a, I guess, a makeover experience where they, give them the whole treatment. You know? They, you you know, they receive them as princesses. They, you know, do the whole, you know, princess set up, change their outfits, and then they take him for pictures. And the whole time, they're, like, blown away. They're excited.

Gustavo Valenzuela [00:57:22]:
They they do, you know, taking in the role of princesses. So Mhmm. That was a a great experience for us and for the kids. So that's my pick of the week. Disney princess.

Charles Max Wood [00:57:36]:
Awesome. Yeah. It's funny because all of my wife and my girls, they all have different princesses that they identify with. And what's funny is is none of them picked the same one.

Gustavo Valenzuela [00:57:46]:
So That is yeah. That's no. That's good. That's good. Because I know my 2 daughters, sometimes they they wanna pick the same one. And, the little one go like, no. You're not. You're Aleph.

Gustavo Valenzuela [00:57:57]:
You can be Aleph.

Charles Max Wood [00:58:00]:
Yeah. That's funny. Yeah. For my wife, it's Aurora Sleeping Beauty. My oldest daughter, it was Cinderella. My middle daughter was Ariel, and my youngest daughter daughter, funny enough, you see all the little girls that wanna be Elsa. She wanted to be Anna. So, anyway

Gustavo Valenzuela [00:58:19]:
Good. The one that

Charles Max Wood [00:58:20]:
didn't have the magic powers or whatever.

Valentino Stoll [00:58:22]:
But, anyway

Charles Max Wood [00:58:23]:
Yeah. Very fun.

Gustavo Valenzuela [00:58:25]:
Yeah. I feel like there's a big marketing and business, lesson that could be learned through that. I think it's have the market partner. Hey, now. Do you think?

Charles Max Wood [00:58:34]:
Yeah. They do. Yeah.

Gustavo Valenzuela [00:58:37]:
In any case, what's the

Charles Max Wood [00:58:38]:
Good deal. If people want to find you online, Gustavo, where do they find you? Or if they wanna hire you because you said you were looking.

Gustavo Valenzuela [00:58:45]:
Yeah. On the market. Yeah. So, Twitter, I'm usually that's where I interact the most. G v g, v as Victor, 1180. And, or yeah. That's pretty much it right now.

Charles Max Wood [00:59:07]:
GV 1180? Correct.

Valentino Stoll [00:59:11]:
K.

Gustavo Valenzuela [00:59:11]:
So it's not Gustavo, my first name for my first letter initials for my name, and then 1180. GV 1180.

Charles Max Wood [00:59:20]:
Awesome. Yeah. I just put that into the, YouTube and Facebook. Yeah. I'm looking too if you're hiring chuck@topendevs.com or email me. Alright. Let's go ahead and wrap it up. Thanks for coming, Gustavo.

Gustavo Valenzuela [00:59:37]:
Thank you. I appreciate you guys.

Charles Max Wood [00:59:39]:
Alright. Till next time, folks. Max out.