Transcript

‍

Run as fast as you can, burn something out, break something down, build something new, adjust whatever you need to adjust. And based on the information that you're gaining and the significance of the results, that will guide you to where you need to go. A year from now, two years from now, having that small team that's maybe a designer, a developer, and a marketing person on one team that can all work collaboratively and have that pace of innovation, it's going to be a phenomenal world.

So now the agent can go on my behalf and find restaurants that are similar, pull them together. I can now have a drop-down list of the place that I want to go to and the kind of meal that I want. It's like, boom, here's a refined list.

Here are the dishes. Ba-ba-ba-ba-ba. I guess we're still in the world where like you need teams of engineers that can glue all these things together.

There's not like one go-to agent that people are deploying that has all this stuff put together. Like we're still in the early days of folks kind of gluing all these tools together to like build custom solutions. Everyone knows JavaScript, but it's like the way you manipulate it, use it and leverage it is the way that you build careers out of this thing or companies.

And so it's kind of the same thing here. It's like you can know these tools, but the method of which you leverage it is what's going to be the fruitful future that you present for yourself. People that I genuinely revere in tech, geeking out about and trying to like desperately know more about.

Everything was around evals, evaluation and how you would evaluate prompts, how you would evaluate information, how you would evaluate the quality of information. This was like a massive, massive push. What's going on, everyone?

We are back with another one. I am literally, I've been home 10 minutes. So I'm super excited to be recording this.

I didn't want to miss this. So as soon as I walked through the door, kissed my wife, kissed my kid, threw them down to the side, and came over here so we could record something real fast. I want to get this episode out.

Hey, I was just in California and you weren't there. I'm just saying. It was easier for you to come find me.

Regardless, let's get through these intros. I am one of your co-hosts. My name is Danny Thompson.

I'm the director of technology at a company called This.Labs and community organizer, conference organizer for the Commit Your Code conference. And I'm Leon Noel, managing director of engineering for a lovely nonprofit called Resilient Coders and community member at 100 hundred devs i am just coming back from the ai engineer world's fair conference in san francisco and it was a phenomenal experience there was a lot of lot of value in that place and i think more so than anything else we need to talk about this okay there was a massive amount of value in that place if you haven't seen the conference if you haven't heard about the conference don't worry about it we're going to cover a lot of the important things that occurred there but we're going to be sharing things in a way that if you it's not i'm not going to cover what was in conference talks they all that is on youtube go check it out i. I highly recommend it, by the way.

There was next level value in that place. Highly recommend it. What I do want to talk about more so is some of the things that probably weren't mentioned on stage or the emphasis wasn't really felt because you weren't there, right?

I learned a lot. You weren't there, right?

I learned a lot. So I'll tell you this. When we got there, my CEO, Tracy Lee, phenomenal human being, she's a rock star.

She said something to me, which I'll be honest, at first I disagreed with. I was kind of resentful a little bit because you're at this next level conference, 3,000 people there. And I'll be honest with you, it literally felt like, just to put this into terms that make sense, if you were to imagine what the world felt like and how people that were shaping this thing, the creation of the internet, how they felt knowing they're creating this thing and they see long term.

They're not seeing the next 12 months. They're not seeing the next 18 months. They're seeing like the next five years, 10 years, what this looks like.

Knowing what they know, that is exactly how it felt attending this thing and being a part of these conversations with these individuals. it felt attending this thing and being a part of these conversations with these individuals like there were individuals in that room that were geniuses of genius level and they're so intelligent to the point where where i'm having these conversations with people that i revere to be geniuses their ideas and the way they talk about it make it feel like feel like no one knows what they're talking about because they're just at such a high level of intelligence and understanding. It just blows your mind away, right?

And I felt privileged to be in these conversations and to have the takeaways. And so I want to share it so that way people can kind of see something on another scale and another level. By the way, this episode is not sponsored by the conference or anything like that.

They have no clue that I'm recording this. But I feel so enlightened by these conversations that I'm like, we have to share this. Yeah.

You think about going back to the beginning of the internet and people had the vision, but they didn't know it could be what it became. And now with the AI boom, people have the vision, but they know what it became and now with the ai boom people have the vision but they know what it can become they've seen how rapidly innovation happened and so now you have these people that you're saying that are operating at that level that also know hey this could change everything right like that deep-seated belief so definitely really curious to hear what what the word on the street is these days yeah it's funny because funny because you walk in thinking you know something about a technology or a tool and you walk out realizing, it's not even that I scratched the surface. I didn't even scratch the surface yet.

But I will say, talking to a lot of these developers, you understand where their mind is in a technical capacity and you understand more so than ever before why some of these companies go to these nine ten and eleven figure marks because for a lot of developers i don't want to blanket everyone so i'm not going to say everyone but a lot of these developers that are so amazingly technical you can truly understand that they don't realize the first thing about business and the ones that understand both sides are the ones that are capitalizing on another level. Like one of the things that was shared a lot, and I kind of jotted these down on a piece of paper so that way I can at least remember some of these in an order that I want. But the one thing that a lot of people had a ton of enthusiasm around and shared a ton of talking points around was this idea around the verbiage of tiny teams and they it's such a big deal that ai engineer warfare made an entire track for the subject because so many people were into it by the way mind you i think they had 18 tracks i could be wrong on that number i think it was 18 tracks which is just mind-blowing um but apparently it's more than double the number of tracks that they had last year but uh they wanted to make it feel like you know if you wanted value you were definitely getting more bang for your buck but it was definitely like hey we're competing with a lot of conversations and you're going to definitely miss a lot which is also appreciative why they shared it all on the internet right because the tickets for this thing were not cheap it was very expensive which also i kind of believe is a reason why you had the quality of the conversation that you had in there because if you didn't know what you're talking about you definitely got priced out for sure that you're you're you're going to be an authority on whatever it know what you're talking about, you definitely got priced out for sure.

You're going to be an authority on whatever it is or you're trying to learn from the authorities when you're spending that kind of money. It was definitely expensive. I believe, if I'm not mistaken, it was like $1,200 for the general admission.

And I believe if you were a leader trying to join leadership tracks or something, it was like $2,000 plus. So pricey, but if I'm honest, you could definitely tell. Now, here's the thing.

I don't know any conference in the world that doesn't have some kind of pain points. There were definitely some pain points that occurred at AI Engineer. But in my opinion, I don't know how they could have avoided them.

engineer but in my opinion i don't know how they could have avoided them they chose the venue because the venue specifically was known for their wi-fi right they're known to have like great wi-fi the venue unfortunately dropped the ball with them there's nothing that i think the conference organizers could have done there because that's on the venue right and so um i can't put blame on them for that for having spotty wi-fi and issues around that because you're going to a place and you're spending ridiculous amounts of money to be in that place for them to take care of that infrastructure what can you do in that regard so i don't hold for any pain point or like growing pains are like man at times it was like genuinely packed i mean you've got so much value and you got 18 tracks you just you'd only imagine that to be a thing at that point right and i mean to be honest if that's a problem that you have that's a pretty good problem like so many people are truly seeing the value in the thing that you're bringing that they're coming out so again i don't really hold that against them uh sure you learn you grow you kind of iterate I think this is only the third year that they've done this. But to be completely honest with you, I don't hold – I personally can't see anything negative to hold them against. And the other problem too, if I'm being real, he even showed a graph of the timeline of people buying tickets.

I think 90% of the tickets were bought two weeks before the event. So it's like if you have 3,000 plus, 4,000 plus people showing up to a thing and 90% of the tickets were bought two to three weeks before the event, there's only so much strategy you can do around that to kind of prepare. I mean, again, I don't hold anything against organizers for that.

I had a blast. I could definitely see areas where they were like, hey, retro, let's improve this next year. But probably choose a different venue or something along those lines based on that or make sure the venue has their things in check.

But I'll be honest. I don't hold a single negative for them. I loved everything that I got to experience.

That's awesome. Definitely put it on the radar for next year. So go back to those little the little teams yeah tiny teams tiny teams i you have to understand number one just got off the plane so the energy drink is energy drinking right now but the other thing too is like i'm filled with all this like yeah ideas and stuff it's got me running so tiny teams um the idea around tiny teams was this there are ai is giving teams especially when they know ai best practices and ways to leverage it to be a small agile team that are building million and million dollar companies in short periods of time like one company that i ran into and we were just geeking out.

They're like the founding engineers and this and that. But they have created a company that is doing like healthcare billing. And it's a team of six.

And they have built this team to where they're literally making $10 million right now. And they've only been around for nine months. Man.

Like the conversations. And what's funny is like one of them has only been a junior dev in one company and after that they joined this as a founding engineer yeah it's like some of these the pace of what you can get to these high dollar amount mmr annual run revenue run rates it's just absurd i know people in the valley are chasing that right now but it's still absurd to think that a team of six can hit a 10 mil run rate in just a few months right where you used to have to slog it out for for years to find that product market fit to hit those types of numbers and every once in a while you had a unicorn but just a number of companies that we hear that are doing it and doing it. Danny, we got to stop the podcast.

This could be our code in time. What are we doing?

We hope you're enjoying the show. We'll get back to it in a second. But first, a word from our sponsors.

Financial literacy is important, especially if you're in a position where you are changing careers and income brackets. More income can create larger financial issues if you don't budget correctly. So this episode is sponsored by Lucas Casares, certified financial planner and founder of Level Up Financial Planning, where he specializes in serving early and mid-career tech professionals to help you take your financial confidence to the next level.

Danny and I have helped thousands of people level up in tech, often going from zero to six-figure careers, and we've seen all kinds of mistakes. Level Up Financial Planning helps you gain clarity on all your finances. This includes prioritizing manager-employer stock, retirement planning, tax planning and preparation, college planning, and as a special deal for our listeners, you get a complimentary meeting with Lucas.

So please visit levelupfinancialplanning.com to find out more. And now back to the episode. Yeah.

You know, it's funny because my wife used to joke like, oh, you're gaming with your friends. And she shared with me a picture of a garage where – the garage where Apple started. And she was like, Apple started with several people that knew each other and working in a garage.

You have developers that are friends of yours and you all are playing the game. What are you doing?

And I used to joke around for a long time like, I don't have a garage. So that's the reason why. But now it's like, I've got a garage like so what is my excuse at this point right like we gotta get something out but yeah so uh the tiny teams that the theory is you can move very fast you can like pivot wherever you need to pivot so like my adage kind of goes to this right fail pivot grow this has been the mantra that kind of rules my life and they're basically living it in real time where it's like, run as fast as you can, burn something out, break something down, build something new, adjust whatever you need to adjust.

And based on the information that you're gaining and the significance of the results, that will guide you to where you need to go. And because they're so tiny, they can work at a pace that maybe some enterprise level organizations can't match. And that's the scaling that's happening.

Like Cursor is a fine example of that. Windsurf is a phenomenal example of this. Like as household names that we're familiar with.

And if you're not familiar with those things, Google it. But I mean, Windsurf, they just got their Series C evaluation. I believe they were fundraised at a 9.9 billion dollar fundraising and cursor did not exist too long ago their team is now the team has grown because you got nine billy you can grow right at that point a couple more yeah yeah we don't need to be a tiny team anymore we can just get there you go windsurf a same example they just got acquired by open ai and so like even that right there tiny team got acquired for billions um unfortunately you know not nine billion three billion you know they missed a couple you know i'm sure i'm sure they're gonna struggle but it just goes to show the impact that you can make now with a tiny team and it's kind of been the thing that i've been seeing where enterprise is a different conversation but it's like what you're able to now achieve as a smaller group is pretty incredible but there's going to be a point where you start expanding and getting to these larger sectors to where it's like okay now we need an accounting team now we need this now we need that accounting team.

Now we need this. Now we need that. You're going to grow regardless.

But it kind of goes to one theory that I've had for a while is that I don't necessarily see AI being a thing that could always be a headcount reduction tool, but it is an amazing productivity tool. And so now if you're a company that has 100 devs, but now you can get the strength of 150, you don't need to lay anybody off. You can expand in areas where maybe you didn't have budget for before and see the impact.

But the best part about it is if that idea becomes a terrible idea, you're not hiring on 50 new people to essentially for a failing idea where you're going to let them go afterwards right but one other thing that was really discussed which is something that i've loved and i've kind of talked about i don't know if i've talked about on this podcast but one issue that has always existed in tech has been prototyping meaning when a team is working on something like ah we don't have time for this or we got to allocate so many resources for this what if i ramp up something like v0 bolt new etc and or lovable you know that's a great one too and i now build out my idea in in an hour as a prototype to then showcase to my product managers this and that they can use it and so it's like a lot more life than the basic figma doc. But now they can experiment, use it, and based on this prototype of functionality, I can now pass this on to UI UX, and they have a better idea of what it is that we're trying to achieve. It's no longer a theory on a Google doc, but it's like, here's the working thing that we're trying to build.

Let's build the design around it now. The wave is incredible. And so even that, like the speed to prototyping has dropped dramatically to where teams can really like run with this the funny thing is that the democratization of it so a lot of companies had hit squads right internal teams that were just super engineers like highly highly capable teams of individuals that can be given an idea spin it up make see if it validated see if it works and some of these were like some of the best people at an org the fact that you can have that same pace of innovation that speed with just a normal team of engineers now and that and maybe and the tools are still getting there right but a year from now two years from now having that small team that's uh maybe a designer a developer and a marketing person on one team that can all work collaboratively and and and have that pace of innovation like it's going to be it's going to be a phenomenal world and i think we're already seeing it right like you're saying with all the small teams that we're seeing now but it's i think it'd be interesting to see what happens when larger corporations give this power to their employees too and they they have the ability to use these tools and the orgs shift to where that becomes how you do business at that company did Did you see Zapier release their guidelines on what everyone has to be?

Everyone, if you haven't read those, you must take a look at it. It kind of explained, it was a chart that explained the competency they expect of different roles at their org when it comes to AI. And it's a good way to kind of wrap your brain around where you stand in terms of are you like antagonistic to the AI or are you fully incorporating it into your workflow?

And so it's really interesting which all the companies have similar policies and we're all on that right hand side of incorporating it day to day and run in full steam. I realize myself, I'm not all the way to the right. I'm like-fourths of the way there and so i gotta i gotta get working on it it's funny because uh even at um there a lot of companies funny enough we're at ai engineer and releasing uh like updates on certain things or even um uh and was it anthropic no No, Google.

DeepMind. They released a new 2.5 Pro update at the conference with Logan Kilpatrick sharing that, which is pretty cool. One big talk at the conference, and we had an episode on this recently, was around Model Context Protocol.

This one, a lot of people were sharing it. And to be honest, a big emphasis was around model context protocol. This one, a lot of people were sharing it.

And to be honest, a big emphasis was around, specifically, we need industry standards that are not company-specific that anybody could use. So the one example given over and over again was how USB-C Type-C for phones became the standard for charging right regardless of the organization now and that's good for the user because now they don't need 30 different chargers these companies don't have to like vendor lock in their charging type it can be universal for everyone but now by having that universal type c connector this opens up this entire market for all these external chargers, these adapters, these ways to plug into things, et cetera. Very similar to what we would require and need for applications as a whole going forward.

Because regardless if you're using TypeScript, JavaScript, PHP, whatever it may be, you should be able to tie into whatever these MCP use cases would be for your application without having to create 30 different layers because this vendor does it one way, this vendor does it another way. Are you seeing people push back against the original MCP kind of architecture, like the original Anthropic stuff?

Yeah, I don't think it was originally pushing back. I think it was originally just so new that people were like, well, what if I went with this idea?

What if I went with that idea?

Et cetera, et cetera. And now they're like, ideas are phenomenal. But if we keep this standard going forward, then it's just great for the industry as a whole.

Especially when we're talking about larger organizations tapping into this thing, smaller organizations. And especially making this standard now that it's so new it just means that we're just going to alleviate a lot of pain points going forward for many years to come because everyone is going to adopt this now and so if that's like the base of what you're building on then it just kind of makes sense going forward yeah did anyone announce any kind of big mcp related stuff at the conference they probably did to be honest, after the first couple sessions that I attended and after that, I started focusing on conversations. They probably did.

And to be honest with you, the rest of this weekend, I'm pretty much dedicating to watching a lot of the sessions and catching up. But I will say that based on the conversations, there were definitely releases that were shared. Like for example, Google's Deep Mind with Google Gemini 2.5 Pro.

That was definitely a release that I got to see. But there's definitely a lot more. There were several other things that were shared a lot of.

One is advancements in AI agents and use cases. Like, for example, AI agents definitely was the talk of the town for pretty much everybody there. And one of the cool things was everyone that I had a conversation with and i recorded a ton of content by the way like i'm talking like i have like probably two three hours worth of recorded content of like interviews and conversations with people that i plan to edit as well um that's probably by the end of this week more than likely that's probably going to be out um just because i want to get it out right away but in in regards to one conversation or question that I ask pretty much everybody, it's like, how are you using AI agents in your day-to-day?

And almost everyone had use cases that were outside of the norm, different. One of them was I actually created an AI agent for all of my online food delivery ordering. So it's like, why do i need a guy to kind of go through it's like i have all this context of my favorite restaurants and i have all this context of some of my favorite meals so now the agent can go on my behalf and find restaurants that are similar pull them together i can now have a drop down list of like the place that i want to go to and the kind of meal that i want so it's no longer showing me the full menu because i have it in a very refined search it's showing me this restaurant has this dish and it's rated this you probably will like this place so it's it's simplified this endless scrolling of trying to find a place he's like i'm in work meetings all day right why do i need to like go through this as i'm having meetings back to back to find this like no let me find this so that way when i finally have my 30 minute window of where i can eat real fast i don't have to spend this time trying to find a place or i don't have to spend this time finally saying okay i've got a moment let me order because it takes forever it's like boom here's the refined list here are the dishes that's a cool use case as far as i'm concerned there was another one that he he has an agent that pulls all his uh financial transactions so like he likes to audit his banking so he's like it pulls it and then it'll automatically qualify this was how much i spent at miscellaneous restaurants this is how much i spent on food ordering this is how much i spent on bills and so i can find areas to save every single month that's that's phenomenal it's fun yeah i think yeah um it's just and that's just the beginning right it's we're gonna have wearables that can listen to you go that was good and then now next time you go to order food it knows what you thought was good like that's gonna be such a a wild reality yeah i i feel like um the agent piece is really interesting i i i feel kind of honestly lost on when i'm definitely on twitter when i'm reading about it it seems it's one of those things where i kind of don't know how to implement it directly into my business use cases in a way that is worth the time.

And when I read a lot of the stuff online, no one's kind of giving a clear path to it. And so in that case, to me, it still feels kind of like hacky, right?

Like I still have to like I'm still figuring all this stuff out. And there are definitely tools and things like that I've played with that were a little bit easier to use versus others, but it still feels like I'm gluing a bunch of stuff together and then like singing, singing a song and a prayer to make it work. And so I'm curious if you saw anything at the conference or just in your regular day-to-day life where it's like a way to build like repeatedly like easy work.

I can't say workflows because that's the bad word when you talk about agents. But something that makes it so that you could literally drop it into something you're doing at this dot. Or are we still in that phase where everyone's still trying to figure out and build?

Oh, at this dot, you know, we're doing a ton in relation to ai uh one big thing that we're doing and to be honest i don't think we sought out to do this originally but due to the sheer number of conversations and requests around this we had to create an offer for this but But essentially, it's an AI service offering where companies are basically saying, we have zero desire to become an AI producer. We don't want to produce this stuff. We don't want to make our own local LLMs and tooling around it, et cetera, et cetera.

We have no desire for that. We just want to use what exists currently in the best way possible, and we don't have a clue how to do it yeah and so a lot of what we're being approached with and conversations with are from like existing clients and new clients strictly around this they're like please bring in some engineers that are ai first and we're like we'll embed some engineers in there we'll get them in that like are very ai friendly and they understand the processes and they'll work on your teams. And essentially they'll pair program with people and it's like, ah, I saw how you did this in Cursor right now.

This was cool. But if you did X, Y, Z instead, our results would be significantly better. It will be far more improved, etc, etc.

And that way, because you can teach people theory all day long, but when you catch them in the real moment and then you give them something applicable that they can do instead, that becomes something that they remember for the long term. And so that's kind of where a lot of that growth will happen internally on teams. And you definitely saw a lot of that at the event as well, where companies were bringing teams, entire teams together.

Matter of fact, a good friend of mine from the Dallas area, I won't mention specifics, but he flew out there. He's a CTO of a company, and he brought out his entire team. I think they were like 11 deep at the conference, and they were all like picking up how can we do this better?

How can we improve this?

What are things that we should have on our radar, et cetera, et cetera?

And so tons of like learning from the team position as well. Yeah, now this is totally not for me. This is for our listeners.

But like what tools are you using when you build a team that's going in to do these like agentic workflows like what what's the technology that is it really depends like uh i mean it really depends like if you're if you're building some form of a chatbot obviously you have to build build it utilizing some kind of LLM, whether local or otherwise. Most of the companies, you're definitely going to use some closed LLM model, right?

So it's not openly training the public, which there's a whole thing around that. You definitely need to have some kind of RAG architecture to make sure you're having there, which was interesting because now there's this whole push of we need to kind of go past basics of RAG and do retrieve and augment kind of advancements. That was an interesting talking point that was happening.

They did a whole track on that one. You definitely need something to where you're adding, which obviously MCP comes in, right?

You need more context besides the basics of an LLM query. see mcp comes in right you need more context besides the basics of an llm search um so you or a llm query sorry so you need to kind of have something around that uh nan is like a great tool to kind of be diving into um there's tons of use cases out there you can make it all proprietary you can use existing agents out there i'll matter of fact i'll give them a shout out even though they didn't ask for it maybe we'll bring them on the podcast now and think about it. Ahmed Awas from Chai.new.

It's from Langbase. You can literally make agents with very little effort and work. They basically create an entire framework around it.

There was also like four other companies doing something similar. I definitely think based off the demos that I saw at least he has the best uh solution for this so if you want to build your own agents this is like a really easy way to kind of dive in deep and kind of jump in there highly recommend checking them out at the very least for like an exploratory uh pathway of like yeah introducing yourself to agentic flows well that makes me feel a lot better because i i know all those things right like i know all those tools and so i i guess we're still in the world where like you need teams of engineers that can glue all these things together there's not like one go-to agent that people are deploying that has all this stuff put together like we're still in the early days of folks kind of gluing all these tools together to like build custom solutions which i guess makes total sense i just didn't know if there was something that i was missing in that world um that makes it easier but yeah i've played with all those things i could see me putting them together to kind of build these like solutions i would say chai.new is probably one that would have a lot of that together but i will say and kind of go into one of your points here is like we know these tools yeah everyone knows javascript but it's like the way you manipulate it use it and leverage it is the way that you build careers out of the same or companies and so it's kind of the same thing here is like you can know these tools but the method of which you leverage it is what's going to be the fruitful future that you present for yourself or build for yourself. Love that.

So tiny teams, what are the, any other big things that were?

Yeah, tiny teams. Oh, I didn't even write this down, but this is like a big one. I've been very, you know, pro for this.

I've been talking about this for a while. Matter of fact, even in our cohort cohort for example um one of my team did this in their project conversational ai and voice ai these are i truly have always believed that this is going to be like the future going forward for a lot of things i think the conference emphasized and proved that more than ever before there are a lot of solutions around this but essentially one company in particular, and I'll give them a shout out to, why not, Twilio. I think Vonage is also doing something similar to this, and Agora.

But essentially, the idea is, we have all these conversational AI agents and bots and use cases that we're building. But getting them to attach a phone is a pretty monotonous, painful process, usually. And so they have tried to build, or not tried, but they have built solutions where it's like a couple lines of code max, where you can basically, whatever conversational AI agent that you've built, you can now attach it to a phone.

And so now it's like, let's say, for example, and I'll give this example because I literally know someone who's built this, but you built a conversational AI agent for a pharmacy. So someone calls in, hey, is my prescription ready?

Hey, this, hey, that. It's a very, very predictable conversation. Predictable conversations are where conversational AI thrives.

Conversations that are unpredictable is where it all falls apart. Prime example, Amazon. conversations that are unpredictable is where it all falls apart prime example uh amazon they created a ai agent for like customer service and it failed terribly because there's so many different areas this conversation can go that is just completely unpredictable to where the ai agent can't thrive but if you were doing it for example a pharmacy what am i conversation is going to be is my medicine ready is it not ready in my insurance issues etc etc like all that is very predictable where that's going to go right and you can even tell like when i go to walgreens and i'm picking up like a prescription or something you can easily tell that they already know where what you're going to say within like the first couple words right and so like they've already checked out mentally they let you run your mouth and like okay cool what's your birthday what's your name what's the name on the prescription like they're already ready to go with their hands on the keyboard and i noticed this the last time i was there because i was picking up a prescription for my son and literally he within the first couple words his fingers went on the numbers he's like what's the date of birth like he was ready to cut me off but he can't because it's his job, right?

And so I think conversational AI is very similar in this respect. And so there was a lot of like solutions built around this. And I think going forward, you're going to find a lot of that growing even more.

Especially because think about it. When you saw like the Metaglasses or you see, for example, Android xr google's version of uh smart glasses you definitely and matter of fact i even saw one dude um he was at first i thought uh he may be like um impaired with his eyesight like an impaired person with bad eyesight and he was just staring off in these very like specialized glasses typing away not even looking at his computer. And I was confused for that for a second.

And so I'm taking it off to look at his computer a little closer and then put it back on. And I was like, oh, he has these glasses that are basically like AR glasses. And he's working on that coding and building something out, which is incredible.

You saw all kinds of neat stuff like that. something out which is incredible right you saw all kinds of neat stuff like that and so i think that like as you're talking back and forth with your smart glasses that's a form of conversation ai you know an assistant version um but essentially like there's so much that can happen within that that you can kind of manipulate and make things happen with um it kind of just shows me like where we're going to be in like a year or two from now. Yeah.

That's phenomenal. And the last thing that I'll say that I felt was making a very big wave to where I saw like people that I genuinely revere in tech geeking out about and trying to like desperately know more about. Everything was around evals, evaluation and how you would evaluate prompts, how you would evaluate information, how you would evaluate the quality of information.

This was like a massive, massive push that I saw a ton of information around or like interest, even like side conversations that I would join into. This was like the core theme of many of them for at least from like day two and day three of the conference. It seemed like this was like a guiding thing.

It's so much so that I even saw a lot of booths at the conference that I talked to on day one. When I went to day two and day three, they adjusted their talking points based on this enthusiasm. Oh, wow.

So if that's not on your radar, it probably should be. Any other big conferences coming up?

You know, there's one I hear people talking about in Dallas called Commit Your Code. Commit Your Code coming up?

You know, there's one I hear people talking about in Dallas called Commit Your Code. Commit Your Code coming up. And I've basically realized after a conversation with somebody at that conference that Commit Your Code is probably going to just end up being a yearly event at this point.

I just need to come to terms with that instead of saying, like, just one more year. I think it's fair to say that this is more than likely to be a yearly event as of right now tickets have been on sale officially as of yesterday two full weeks and we have sold more tickets than we had attendees for last year oh wow so we've already crossed that threshold i was worried if we're going to get to that threshold we've crossed it so now we know for sure that there will officially be way more people there this year than there were last year which is is great. It's also terrifying.

So it's good to have growth. But then, of course, you start wondering, can you sustain the growth in a way that people still feel the same way they felt about it the first year?

I'm fairly confident we will, because we just do things different. It's just different than a regular tech conference. things different.

It's just different than a regular tech conference. And so I think even as I was at AI Engineer, there were things that I absolutely loved that the way that they implemented. I know for a fact that there's going to be a completely different implementation than what we do.

And they did things that we probably would never do. And we're going to do things that they probably will never do. Because's just not it's just a completely different feel vibe and mission with commit your code versus any other tech conference and you know i think it's just we just i just see conferences being different i think that's why the quality of the experience is different for people and you know even from the speaker perspective it's very different and so um i think i think it'll be good for that all right it's time for everyone's favorite part of the podcast, including me, to be honest with you.

It's Ask Danny and Leon a Question. And if you want your question to be answered on the podcast, submit it in the email in the description shown us below. I believe it's dannyandleonspodcast.gmail.com.

Someone from the team will look at the question. If it's worth asking on the podcast, we'll absolutely respond to it and ask it. Or I'll be honest we've responded to a lot of them just directly one-on-one in the email response i'm not saying that we'll always do that not trying to set that standard but you know there are moments where i find myself like in between meetings or something along those lines where i may just respond to it because it's someone from the team shared it so um highly recommend dropping it daniel leon's podcast at gmail.com all right leon what's today's question all right my resume is like an 80 for the ats applicant tracking system but not once have i applied and not gotten that we decided to not move forward not sure why wait not once have they got they haven't they had never gotten past submitting the resume they keep getting the we decide not to move forward message there's like a thousand potential reasons for that um the ats is only really measuring if you're matching the keywords in the way that you need to it's not measuring the quality of your bullet points or the responses or anything like that um and as you know as the adage goes if it's an 80 there's still room for improvement right but uh i mean if it even at 100 is like well it could probably be you know finessed in this way or that way etc etc cetera, et cetera.

But I would say the strategies that I often do around resumes when it comes to my calls, a lot I will not share publicly because it depends on the person. Also, a lot I won't share publicly because if I'm being real, some folks literally steal the advice and put it behind a paywall. And I don't really want people doing that.

So I save them for like more intimate like Twitter spaces or conversations around that. But some of the strategies that you need to be thinking about is how is my resume setting me up before I'm in the conversation and how is it setting me up once I am in the conversation?

And the way that you do that is audit your resume. Like read your resume as if you were a stranger and then say, okay, I've read this. What are my takeaways?

What are the value points?

What are we having a conversation about?

That would be some of the things that I'm thinking about there. It's like, okay, Leon gave me his resume. If I were to now call Leon on the phone, what is my hope that Leon is asking me based on this resume?

Or what is Leon hoping that I ask him rather?

That should be your thinking around it. Like what is the strategy around this?

It can't just be like, all right, resume is a checklist on my to-do list. I just need to knock this out as quickly as possible. One prime example, last night as I was looking at my Discord, someone submitted a resume in the resume review channel.

We have one in my Discord, right?

And volunteers go out of their way to kind of respond give feedback and help right this resume was written by somebody who is in like an academic position these are people that i generally would i think the general public would think like surely they have to know a little bit more than the average when it comes to like submitting their resume and making their resume this thing is three pages long and there's like absolutely nothing on here. But the thing is, this person is a manager. And so that's often what it comes back to.

It's like, for example, oh, I'm so glad I just thought about this. When it comes to a role, I think we've talked about this on the podcast before if I'm being honest. But when it comes to a role, I think we've talked about this on the podcast before if I'm being honest, but when it comes to a role, there are certain things that I assume are associated with that role.

One prime example that I always give is the same example every time because I feel like no matter where you are, you can understand this. I want you for a second to listen to this and just pause to see what your response is. But when it comes to the position of a coffee barista, what are things that you assume are associated with that role?

Pause it right now and just verbally think out loud. When you think of a coffee barista, just do this exercise because I'm telling you it's going to change things for you. Pause this podcast right now and think about what do you normally think is associated with this role.

Hopefully you paused. If you didn't, you're not going to pretty much get the benefits of this exercise. But some of the things that you would associate with the role is making coffee, maybe serving customers, maybe even taking coffee to their tables.

Maybe it's unloading delivery trucks for the coffee beans and whatever you need to make the coffee with, right?

These are normal things that you would associate with this. Why would you ever make this a bullet point on your resume?

It makes no sense. You know they know that you made coffee. You know that they already assume that you served customers.

Why would you ever tell them that as a bullet point?

So in a software development position, what are some of the things that I would assume that you do?

Maybe maintain code, build out a homepage, write unit tests, something along those lines, right?

Why would you ever tell them that?

Why would you tell them that you are a part of code reviews?

That's the most basic part of our job. Why would you say, yeah, I attended code reviews, I was in code reviews?

If you wrote code on a team, I would assume you were in a code review somewhere. Why would you ever write that as a bullet point?

And by the way, if those are one of your bullet points right now, take it off. The only exception to this one is unit testing because if you walked on a team that had like no unit test or maybe 10 code coverage and you were able to like meet business requirements and build up to like 60 70 that's a great bullet point because you just show and you have all those metrics to show that growth of you implementing a unit testing framework for your team right i think that's great outside of that you should never have that on your resume. So why would you write that?

How does that even set you up for a conversation?

If they already assume that for the role, what question would they even ask you?

How are your code reviews?

That's not a question you want. Yeah. Go for it, Leo.

I think 80% match is of the yesteryears i've been talking to a couple recruiters recently and the new ats systems that they're using are kind of garbage at the moment because they're all ai infused ai powered and they're looking for exact keyword matches and i think it's actually the ats is getting worse in terms of like how it's filtering because the number of applications have gone up because people have auto appliers so the ats's are kind of trying to be like lockstep being more scrupulous and like looking for more details and so if you're at 80 i think that's a clear you got to get it to 100 like that's the game we're playing these days and it's looking for exact keyword matches for stuff from the from the job posting. So if you don't have exact keyword match from the job posting, there's still weird stuff on your resume. 100% understand why you're not getting pushed to the next round.

It's just it's just a different game being played now and then again the stuff danny's bringing up of having actionable stuff on your resume true xyz format skipping the fluff and stuff that like they already know about your role is really important and the last thing we always say is that you just got to get a human to actually see it and so networking your way in after you do apply and put those resumes in is always going to help because you can actually explain the things that are there to someone that can actually understand well hopefully that helped y'all it's been real it's been fun see you on the next one goodbye everybody peace

‍