The "glasslets" concept is a clever strategy to lower the barrier for AR creation, but the real challenge remains delivering these advanced features within the physical constraints of daily-wear glasses. It is a polished developer-first vision that now faces the ultimate test of hardware execution.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
I Confronted Maverick AI Pro Glasses Tech Lead With Your Final QuestionsAdded:
So, the world's first eyetracking glasses and their Kickstarter campaign in just 3 days from now. And I got another chance to interview the CEO and the head of the software development to ask questions about SDK and all the other questions you guys posted in the comments. So, we're going to go over some crazy stuff here from being able to vocally ask the glasses to create apps for you on the spot. What native apps will launch with the glasses, progressive lens options, and then a bunch of other design choices, other features and possibilities for how anyone could make apps for these glasses and how they can be used and all the different things that they'll come with on launch. But first, if you appreciate me taking the time to film this video in the middle of the night in Croatia and Europe right now because of the time difference and then edit the entire thing in like a couple hours because this has to go up tomorrow cuz this campaign is ending so soon. All I ask is that you subscribe and as a thank you, I'm going to try to keep helping you guys make the most informed decisions that you can and ask all the questions that I can to all these smart glasses companies. Without further ado, let's get into it.
>> Aron is our uh VP of software and AI.
Um, and so he's in charge of the team responsible for things like obviously the software, but also the SDK and apps and so forth. Yeah, great chance to have a have a chat on those topics. I know there's a few questions that have come up. Um, so we can certainly get into those, but yeah, just appreciate you taking the time and and doing another another Q&A.
>> Yeah, of course. Seems to be a lot of interest in them.
>> Hey, it's great to be back with you, Stephen. It's really awesome to have a chance to chat again. Obviously, the campaign has been going fantastically well. We're really really excited by uh all of the engagement and uh not just in the comments on the Kickstarter, but we've actually opened the discord now as well. And so there's a lot of chatter back and forth on there uh including on the development tools and the SDK. Uh and as you and I had talked about previously, one of the things that we think is really really important with these glasses is uh to provide as much access to developers to build the software that they would like to see on the platform. Right? We'll obviously have our apps and we'll have uh the experiences that we'll provide to customers, but we think the most exciting thing is that other people can build whatever apps they wish for the glasses or that they can add support for the glasses into their existing apps. Uh both of those are you know exciting of course and then we also have this concept of glassletitlets which is the notion that uh without being a developer you can ask the AI to write a little app for your glasses and it will do that and so we'll get into that in just a minute because that's super super cool but uh maybe it's time to introduce uh one of the most important people in this whole process and that is Iran Friedel who is our VP of software engineering and AI. I at every site. Um he's joined us from our HQ. Um and so if it's okay, I'd love to throw to him and let him introduce the SDK and talk about what's possible.
>> Great. Thank you, Dave.
Um well the SDK actually is was built on top of the Maverick Sport SDK concept where actually the glasses since they run embedded the operating system. We actually put all the logic and rendering rendering engine within the SDK which runs on the mobile application and we expose the user really high level modern NPI that actually encapsulate all the glasses capability in a really high level uh common mostly common programming languages like Kotlin or Swift. uh this enable you actually to take your existing application or develop a new one with really known IDE like Android Studio Xcode and just add our libraries as a regular uh library and just develop your own application.
Actually according to our experience it's going to take you about I think 10 minutes until you integrate the SDK and actually see information on the glasses.
really really easy, really straightforward. You don't have to learn any new language or technology.
>> That's great.
>> We're going to release the SDK in the coming uh day or two, hopefully even tomorrow. It's alpha version of the SDK.
Then we're going to have publicly accessible the SDK portal, the SDK libraries and samples and mainly the simulator when you actually can experience the SDK and the glasses had without having the glasses.
So developer can you know uh get to know the SDK and when they will receive the glasses they already have the application to use. So it's really really easy. I was just going to add there that I'm not a developer by any stretch and uh Aron likes to joke that that's very much the case but um with the tools that are available even you know not really real developers can still get in there and write some code um and we'll have access to obviously all the AI um assistance uh things like MTP servers and so on as well. So that won't be live for tomorrow for the initial alpha release, but um all of this we're bringing uh to make development for the the glasses as easy as possible. Um it's worth noting uh that the release that's the that we're talking about tomorrow, the alpha release. Um please understand this is a very very early release. We're still some months away from having the final product. So it is very very early days.
You won't see all the functionality. Um and I know some folks have been concerned that if it's not, you know, if it's not being released tomorrow, oh, it's not coming. That's not the case at all. It's just we're doing this in a staged manner. Um, but we are really excited to get the feedback from the community on what they do see. So, uh, if you are a developer and you can have a chance to play with it, we'd love to get that feedback and then roll that feedback back into what we ultimately ship as the final product. And although it's a alpha release of the SDK, actually exposes most of the features, okay, where you can draw 2D graphics, 3D graphics, handle the line of sight sensors, receive all the glasses events like touch, proximity, uh control the brightness, the auto brightness functionalities, display most of the things that the glasses support.
Of course, the next releases like the beta release, the release candidate will have more features like another microphone support, second microphone, uh, eye tracking, full API, and more. On the SDK portal we're going to launch, we're going to have a road map page where you can see all the future capabilities we're going to add in the future version. One thing I was thinking we could do uh if you'd like is I can share a screenshot of the simulator and perhaps Aran could talk through some of the functionality that's there and explain what's coming if that's cool. I just tried to do a showing but it said I had to request that. So >> yeah I can share also the part of the SDK portal which we are going to release.
>> Yeah.
>> Perfect. So this is the SDK portal. It's still not uh public but this this is going to be the uh what we are planning to launch with a bit modification. It's an alpha release. So you have here all the quick start guide.
Okay, it's really really easy. It's one page. It's actually when you follow this, you can, you know, add a HUD to your application just in like 10 minutes. Okay, it's really really easy.
This is how you display information on the glasses. Really, really high level API. You can inherit from a screen, virtual screen, add a text, some color, some positioning, and then you have a text on the glasses. So the SDK encapsulate everything. Encapsulate the Bluetooth uh up channel, the downream channel, everything. The Bluetooth here, since we using B is really, really efficient channel. We made a lot of efforts to take all the information and transmit it to the glasses in a really really high efficient uh B stream stream because all of the information actually transmit via B. So we don't need Wi-Fi.
We don't need a high consuming energy uh protocols.
This is the reason the glasses can actually work more than eight hours with single charge display always on full brightness. No issues here. In addition to the quick start, you have all the information of the supporting uh operating system, general architecture of the SDK, how to set up everything.
We support both uh native iOS application, Android applications, also multiplatform applications. Future versions will also support where OS and Apple Watch. We have guides for each of the uh capabilities we can use with the glasses communication display inertial sensors the AI vision sensor which is the gives you camera okay for the AI engine to work with microphone eye tracker which we have part of the documentation here but it's not going to be released in the alpha version libraries of course the simulator and and our UI kit which which will help you um render information on the glasses.
You can see part of the elements we can draw on the glasses which is vector graphics, text, media, images, fonts, geometry, 3D objects.
Uh again really really really easy to use. The line of sight is really important part of the SDK. Actually, we are pretty much the only MCU based glasses that can actually render 3D information. So, the SDK easily enable you to generate 3D graphics, 3D objects like circles, cubes, like you have in the SDK sample. We can display rotating earth on the glasses. Okay. you can position elements in the virtual world.
Okay, they are stabilized and the SDK enable uh all of this. Of course, our main goal of the alpha release is actually to get the community feedback.
So, we are having a a discord channel which developer can join. We have a GitHub issue which is more like a a developer communication portal. We can use submit issues, feature request and of course we have uh many ways to contact us if it's through our website or the GitHub or developer portal which you can actually ask question get information and help us getting the SDK better. We want to improve all the time.
We're really open to the community uh requests. Uh Dave, you would like to share the simulator screen you have? Uh All right. Can you see that?
>> That's cool.
>> Okay. So, the simulator actually it's uh run on the it's running on the web browser. It it's enable your application.
Okay. To connect to this virtual environment and actually this virtual environment implements all the Bluetooth protocol we actually have on the glasses. Here you see for example 3D objects flying on the world in the world around you which actually if you're going to wear the glasses you're going to see it actually floating in the real world. So this simulates this the simulator actually help you develop because first of all you don't want to put your glasses on the head all the time while developing.
On the other hand, if you still didn't get your glasses, you must have a way to develop. So, this helps you. You can simulate here the touch event, the sensors, the eye tracker when time comes and actually everything the glasses physical sensor let you let you do.
>> So, this will be releasing tomorrow as well.
>> Yes, of course.
>> Yep. Amazing.
>> Yeah, it looks really cool.
>> Yeah, we had a we had a lot of discussion in the discord. I'm sure there'll be a whole lot more coming uh on just you know getting a hold of this and starting to build out some some apps. Um we're excited to get it out in people's hands for sure. Um but also we're excited to get some feedback from the development community on what sort of things they want to develop for and uh how we can make that easier in the SDK as well. So um in the Maverick Sport SDK for example or the the dev tools anyway, we have um AI tools and and an MCP server to help with that. Um, I think that's probably going to be very useful for MAV AI and AI pro development as well. We had some people in the in the discord asking about various AI engines and plugging into that. So there's a lot of discussion around that.
So I think uh AI development will also be a huge part of the the process here for most most folks certainly a lot of folks.
>> So yeah this one for example it's AI generated with the AI here's the MCP server generate a nice welcome screen that this is what you did. So it's as easy as it get. It's it's really really easy. So actually we're going to take this concept and integrate it within our companion app where you can ask our application, show me on the glasses world clock in Taiwan and it just it will do it. Okay, we'll do it with cool graphics and you can tell him no, I want it in red, not in blue. And actually it's really really interactive and um it's very cool because the SDK supports all the graphic capabilities and they can generate.
>> I have a different one here on just to give some other examples just showing connecting to the the simulator.
>> There's so on the left side you're going to see the the sample application we're going to ship with the SDK alpha version. It's going to be open source.
So everyone uh can see the code of the application to see how uh to use the SDK. So here you actually see the uh opensource application that you're going to launch with the uh simulator that actually every developer can download in the developer portal and just compile it, use it, change it, modify it and learn how to use the SDK.
>> And then I've got one more here I can show. So this is our live AI app that we're using for demonstrations right now.
That's swiping on the on the touchpad.
Demonstration of how swiping works here. You can see by the way that actually you already saw the glasses but you know it's a full color really bright display. Um in order for example to generate this uh had Actually, it's about two lines of code. Just put a text with the, you know, with the positioning and you have the you have the information on the glasses.
>> And then, as you saw, Stephen, when you have the eye tracking, you can use the the uh the eye tracking to swipe left and right uh by looking at the edges of those cards, wait till the blue line expands, and then and then the swiping works too. So swiping, clicking, uh scrolling, all of these uh types of functionality supported with the eye tracking as well. Again, not in this alpha version, but in the the next next version to come and certainly in the final product, >> we think the glasses, you know, you need to control them with the glasses and themselves and what do you have with you? We don't we don't know want, for example, external accessories to control the glasses with. So in that way you have the touchpad and you have the eye tracker and you have line of sight and the glasses fills you. For example you could look up and down and it will open a menu. We don't need physical controls you know that you have to wear all the time that you need to charge that you need to take with you that you need to pair to your phone or to your glasses.
We need we want everything to be simple.
We want the user just to, you know, wear the glasses as regular glasses they wear every day and just to use them.
>> So, it looks like a lot of the craziest possibilities are just stuff that everyday people can make pretty low effort based on all this.
>> I think so. I think that the tools here I mean obviously these are very developer focused tools uh the simulator in particular but uh I can tell from my own experience that the advent of AI has um created much more opportunity for non-professional developers to get in there and write write some code um or at least write apps um but we're going one step further and we're going to make it so that anyone can build very small little you know glasslets we call them um and perhaps we can talk about that for a moment because I think this is a pretty big innovation for folks Um the concept of glasslets is that and you can think of it as you know applets.
So small apps but for glasses so we call them glassletits. Uh the idea is that you know for um sort of incremental functionality like a world time clock or a little chess game in in the viewfinder or something that is uh probably not a fully you know uh little apps glasslets we call them that you can ask the AI to make for you and it will know all of the characteristics of the glasses. it'll understand exactly how to how to um position, you know, a a a visual in your line of sight. And so you'll be able to say things like the example that Aran just gave, which is, hey, show me a world time clock with, you know, these two cities, uh, and make it pink and blue and, you know, whatever you may may like. Um, so we think that for a base level of functionality, there's just an opportunity for people to to uh create their own and you won't need to write any code at all. That's all fully within the AI experience. So we're looking forward to having folks experiment with that. Um Aran, do you want to add any comment? I mean that's a very simplistic view, but um what what's your sort of developer um over overview on that?
>> Yeah. Well, we're going to launch just with the Maverick AI glasses themselves.
We're going to launch a companion application. Okay. The comp application actually will enable users to use the glasses without any dedicated application or third party application.
So we are going to support all the user with our own developed application that they will have access to AI and to all glasses functionality within this application. We're going to have the glassless feature actually where for example you're going out there, you're not a developer. You just have our companion application. You can have all the information to ask the weather to to do navigation to do whatever you like.
You don't need any external uh application or something. And then for example, you want to play Tetris and we don't support Tetris in our uh companion application. So you're going to tell the glasses, hey, I want to play Tetris. Create me Tetris game. And actually the glass slips engine will create you a Tetris game fully functional. And if you don't like the colors, you can ask him, "No, I like the colors in red. Add some sounds, do things like, you know, like vibe coding, but you're going to talk to the glasses and receive all what you want on the head-up display."
>> Yeah.
>> Fully interactive.
This isn't like typing it in to like an app. This is like actually speaking and then like the microphone interprets it in the glasses and does all that.
>> Exactly.
>> Wow.
>> It actually takes vibe coding into regular people. Takes vibe coding into non-developers on the go outside. You can vibe code with your voice when you're going outside, you know, walking outside on the street.
>> It's absolutely bonkers. And first time I saw this, I was like, "Yes, there's no way. There's no way." And then I literally watched as it built me a Tetris game like right there in front of me and then bang, there it was. I was like, "This is this is crazy, right?" So I think I think folks are going to have a ton of fun with this. Uh excited to see what what what gets built. But um yeah, when the system knows like all the parameters, it knows exactly what the um you know was required to put an app on the screen uh and you feed all that to the AI engine and then you just let people ask for hey make me X. Um turns out there's a lot you can do. Um and so obviously we're going to keep innovating on this. We're going to keep evolving it. Um it is very early days today but what I've seen already is is is kind of bonkers. So we're really excited to demo that very very soon. Uh maybe it's worth just commenting also uh on on this topic uh AI. We get a lot of questions about um you know support for AI and hey I want to use my existing AI whatever it may be is that possible um and I I wrote a post in the discord talking about how the AI engines uh for the most part don't allow you to take your existing account and then apply it to a different you know surface such as um glasses.
uh if they were to add support for the glasses SDK for example in the chat GPT app or the claude app then story but uh until that happens um they generally don't let you do that now um we will have an AI experience that will be native and built in >> um we are expecting to to allow people to add their um their API key if they're if they have you know an account with the AI engines that has AI API keys um but as I also put in the the post that will come with some different functionality or some limitations because uh some of the core functionality we're using right now is not supported by all the AI engines. So um obviously things are moving super super fast and who knows what will be the case in you know 2 3 4 months time.
Um a lot is happening every every week right now but um we're trying our best to make it as open as possible. We know that there's a ton of interest in people using whatever tools and and and AI systems they have already. Uh and our goal is to make that accessible as best we can. Um Aran, do you want to add any comment?
>> I think the basic experience we're going to launch will be sufficient for most of the users. For example, if you have met glasses, you don't want to, you know, switch the AI engine of the meta glasses, right? I think for h some very dedicated specific use cases I guess several developers would like to add more functionalities or develop something. So in that case we're going to uh try to give them this option. First of all they can develop their own application if they want a totally different use case.
Even if you're going to B2B or B2B TOC and you have a cooperation with know some company you want to develop I know a touring application or a warehouse management application. So actually that I I guess they would like to develop their own application rather than change our stock application.
So in general I think our base capabilities would be sufficient for you know 99% of the users. We have as dev said we're going to have the option to use external AI engine but it's going to be more limited and of course developers can develop their own dedicated application for really specific use cases. We're open to everything. Probably the biggest question we're getting here is, and it's a totally valid one, by the way, is, "Hey, I've already built up a lot of history, a lot of context in an AI engine, right? It knows a lot about me.
It knows a lot about my preferences. Can I just, you know, bring that over and use it with my glasses?" Uh, it makes complete sense and I totally get it. Uh, the only problem is most of the AI engines don't allow for that. Um, and my personal opinion, just, you know, speaking off the cuff here, is they're going to have to figure this out.
because it is absolutely going to be the case moving forward that everything's going to plug into AI and if you're going to bet on let's say you have a Grock account um as one of our users mentioned in the in the Discord you know if you're building up a whole bunch of context and history and you know background on you in Grock then you're going to want that to be portable right you're going to want that to be access accessible by your car and by your oven and by your glasses and by your watch and phone and all these things right Um, unless these guys are going to put their tech into all of those things or in this case put our SDK into their app, uh, they're going to have to find some way to share, you know, share this soul information, right? If you're an openclaw user, you'd be familiar with soul.md. It's basically what this is.
So, anyway, that's an unsolved problem.
Uh, I'm sure there's a bunch of people working on it right now. Uh in the meantime, we will do all we can to support whatever is possible, but uh folks should just be aware that right now there isn't the means on most of the AI engines to just plug in and use what you've already got either your existing account or the context that that comes with that. So um but that's okay. We'll work with whatever we can and you know as I said this is a fastmoving pace sorry fast moving space uh and the pace of which is only accelerating. So we'll try and our best to keep up. You'll keep up. you'll see us keep on innovating as well. Um, but whatever is possible to support, we'll do our best to support.
So, our goal is not to to be closed. Our goal is to be open, but we want to provide the best possible experience for users as well. And there's just some functionality such as the the voice and interaction, you know, conversational interaction that is not possible on some of the AI engines, but is on others. So, that's where we're putting our investment right now of time, but uh, yeah, we're looking forward to seeing what comes. I think they do this on purpose because they don't want I think OpenAI for example or cloud tropic or whatever they won't give you the information because they don't want you to transit to the other one they're competitors.
>> No, I was saying I think that that was most of the topics I think on the dev and AI side.
>> A couple like commenter submitted questions. Demitter 3 asked uh which native applications will be launched on the first day.
>> Yeah. So as I said the the companion application which actually uh covers most of the use cases of the glasses we're going is going to launch uh on day one. It will enable you actually talk to the glasses and use the glasses sensors um with all their functionalities. You can ask the weather the stocks. We have our own uh AI agents uh back end actually takes the information, take the sensors and take the user uh requests and actually uh process them and we give you uh the answers both visual and vocal. Uh we do plan to take some of the Maverick sport applications and to transit them into the Maverick AI uh glasses. Okay. It's going to take time.
Not sure it's going to be on day one, but we will do it along the road map.
It's part of our release road map. Of course, we're open always to partners.
We have several partners of Mavic Sport that will probably uh transit their application to MAVIC AI and we still gathering more information from the community in the Discord channel uh and the developer portal that we're going to launch about more uh community requests >> things like translation, transcription, transcribing um I was going to say there's a bunch of a bunch of functionality that people will have on day one. There's a bunch that will come soon after. Um and then part of what we're doing on the business side is working with partners who are interested in building applications for the glasses. Um and in some cases they won't be day one but you know fairly close. Uh and in fact I'm uh getting on a plane to Europe tonight to meet with some of those partners to talk about opportunities with some uh European companies for um software for the glasses as well. But um we can't say too much about that just yet. But uh obviously our goal is to ensure that there's a an ecosystem of a developers of B partners. Um there's a a base level of apps that we can provide of course ourselves and we'll keep keep um supporting those and rolling those out.
And then of course the glasslets as well for folks that have um you know the the want to ask you know the AI to make um little apps for their glasses as well.
This a comp application it's not just a single purpose application. It's a multi-purpose application. The same application is going to have you know going to have weather talk generally I conversation both translation transcription. So everything is going to be in the same uh comp application. So it's going to be like a Maverick AI pro operating system that actually runs and do most of the thing you actually need specific use cases from our develop from you know in-house development or external partners. We're going to have more applications going to have also many application within many more capabilities like translation transcription checking the weather stores or any other uh tooling. You can search for places and you can actually navigate to them.
So imagine yourself walking in Paris or London and you're gonna you can tell the glasses search coffee shop around me.
You're going to see the information and tell them navigate to the coffee shop and then you don't have to take your phone, you know, to hold your phone with Google Maps or whatever. Can just go with the glasses navigate. You can wander around the street. You can see something. You can see sign. You can say okay what is this sign? What does it mean? Can you translate it to me during the navigation? So you get everything in a single package in a single application >> and again for a dedicated use case like I don't know playing golf probably we will create a dedicated application which actually targeting golfers >> okay or we might have partners from the sport domain or I don't know wind surfing probably they will want dedicated application that they can create or we can create for them uh using the SDK but on day one going to get a lot of capabilities built in as a single companion application which be will be like the operating system of the glo and then you mentioned earlier you were considering like transferring some maverick sports apps over do you like have an example of those because I know there's like like a golf one I think there's like a health one >> yeah we have we have golf we have translation which actually is going to be uh built in within the company application We have sport application which actually it's used for cyclists or runners that want actually to measure their performance. You know they connect to heart rate to bike sensor like power meters or uh heart rate monitors uh more like a you know like straa application or G application similar to this. Uh the advantage of advantage of this application of course that you can see all the information on the head-up display. You don't have to take your wash and to look at the measurement and while you're running everything and jumping actually and you can see anything. Everything's going to be display on the HUD. We have a driving application for example you can take your ebike and navigate throughout the city then to put your phone on the handlebar and to look at Google whatever we do even use. We can use Google navigation engine within our application, but you see all the information on the HUD. It's very safe.
You don't have to take your eyes off the road to the phone. You just look at the road and see the information in front of you.
>> Yeah, that'll be some people get really excited about that.
Yeah, it's uh so the main applications we you know we use actually handsree hands-free outdoor applications actually you want to go outside you want to see the world you need additional information you don't want to take your phone out of the pocket or your hands are busy and you can't do that do this >> so actually this is what the use cases we're trying to solve I personally ride the road bikes and I have like tens thousands of kilometers with the Maverick Sport.
>> Wow.
>> Actually, it's it's hard for me not to ride with them, you know. I see for imagine yourself riding like a 70 kph downhill in some really in on the road and it's really hard for you to to to see all the turns and the curves and you have the map in front of you on the road. So, it's really really easy. you feel really safe and we're connected to the for example to the Garmin radar.
So I know where there is a car behind me coming so I can go to their side of the road because I have the Garmin rather connected and I see the cars coming behind me on the glasses display.
It's really really helpful. I don't have to take my eyes off the road and to look, you know, backward to see if something is coming.
>> I think the golf app is another great example of um, you know, when there's some really rich functionality that we can offer. Uh, I know I think seven, you've seen the golf app in Mav Sport, right, where you've got >> perfectly clear, nothing on the screen looking straight ahead. You look up and you get your score and the weather and the time and then you look down and you have an overhead view of that hole and you can see where the bunkers are and see where the green is and you can um see the distance to the green and so forth. Um that's a great example of like a really rich experience. It's a combination of line of sight. It's true AR. You know, we overlay when you're looking at the green, we overlay the distance to the green so you can make club selection. um things like that are a great example of actually using the functionality and features of the glasses in a very rich way for a particular scenario that wouldn't be as part of our central sort of core app.
That would be a dedicated app and uh of course third parties could add that support to their app as well. So uh so you'll see us obviously release and port over those over the course of time. Um but I think you'll see also some third party partners building some pretty nifty scenarios uh into their apps as well. So, we're excited to bring those out as well.
>> So, someone had a question about the pricing that they were confused about.
Said Kickstarter says 389 and 325 for normal and then some of the advertising says 2.99. That was like an early bird special.
>> That was the VIP. Yeah, that was the the pre-release uh pricing.
>> So, now it's 325 for the Maverick AI. So that's just you still get all the cool apps but without eyetracking and it's 3.89 for the AI Pro which is all the same stuff but with eyetracking. So like I'm on the the website right now and it says that there is 19 of 200 available left for the Maverick AI and then 22 out of 2,300 left for the AI Pro. So like those deals are almost done.
>> I think we'll we'll certainly make make more available if needs be. I don't think men turn people away.
>> Okay.
>> So, I wouldn't worry about that too much other than get in quick because of course it's only 4 days left and time is running out.
>> Yeah.
>> Um but no, these are great prices and just to give people clarity. So, you know, the AI Pro which is exactly the same as the AI but has the eye tracking for UI navigation and for AI um um propless querying and so forth. Uh that will be $5.99 at retail. Uh, and then the version without the eye tracking, which is just the Maverick AI, will be $4.99 at retail. So, these are great prices. 325 for the Mav AI and 3.89 for the AI Pro. Uh, both of those are in the order of 35% off. So, very, very attractive pricing and obviously we're keen to reward the early backers, but uh, yeah, it's time is running out, so time to get in. Uh someone asked to discuss the single ear speaker versus like consumers are like used to getting dual ear speakers for smart glasses. So >> Yep. Yeah.
>> Yeah. Very happy to. So um this has been a question we've had a lot of uh you know discussion about. Um and by the way this is a topic that very reasonable people can debate you know quite openly and you can see all sides of the conversation right. Um, we've made a decision that we are optimizing for weight and we're optimizing for battery life and we're optimizing for comfort.
We did a lot of analysis of the market and we saw how people use glasses uh that are out there today and without any doubt there are a percentage of people that buy for example the Meta Ray-B bands uh and they take a ton of photos and a ton of video and they're very heavy users of that and that's great but overwhelmingly we had incredible feedback that a lot of people buy those glasses take a few photos and then don't take photos again or take them very very infrequently. Um, and so yes, there are customers who buy them for that reason, but a lot of people are buying them because they are a smart version of Ray-B bands, and they want a good-looking pair of glasses. So, uh, we took the decision that for this generation of glasses, we're going to optimize around the weight, the battery life, the comfort, and the price, but also really importantly, the time to market, right? We're not going to suddenly go and change a design we had uh and add in super high resolution cameras and you know multiple speakers and whatnot and then delay things by months and months and months and months and months. Uh we wanted to get the glasses to market. But really the design decision made here was these are an AI smart pair of glasses that you will wear dayto-day. You wear them as your sunglasses. You wear them while you're driving or playing golf or running or whatever it may be. Um, and so for people who want to listen to great music, they'll probably use their ear AirPods or earbuds anyway. Um, but we did want to have the ability for audio feedback on things like AI queries, which you don't need to hear in stereo.
That's just something you can hear uh because it's text being converted to voice. Uh, likewise with a camera, a little bit different with the camera actually. We thought that uh where the world was going right now is um people I think are tending away from having a camera in their glasses as a as a default. I think for those that are sort of early adopters and a very tech you know not I won't say the tech nerds cuz I'm one of them but um you know the folks that are early adopters they're comfortable having glasses. I think the majority of the world is not there yet.
And I think we're going to find especially in places like Europe where they tend to be uh a little bit more aggressive on regulation and that sort of thing, we're going to find that uh they're going to be clamping down on putting cameras and glasses. They're going to require, you know, bright lights that indicate you're being recorded. There's a social concern there about being recorded. And then the example I love to use and people have been quoting me in the in the discord which is we want to turn on scenarios where if you're taking videos of or or video or or photos it's a little bit creepy but the AI scenario is totally fine right and the example I use commonly is wow I love that woman's dress where would I find it that's a very reasonable AI query it gets pretty creepy though when you're taking photos of women in the street right and that's just one example where we think that if you really care about that scenario in glasses, then there's lots of glasses out there that have very high resolution cameras, rock on. We encourage you to go check those out. But if what you're looking for is all day wear, lightweight, battery life, and the ability to have a great AI experience, then that's what we're optimizing for.
So, you know, we've made those decisions. Um there's there's healthy debate in the Discord about it. No problem at all. We welcome that. Um and we'll see how they use. We'll get feedback from users once the glasses are out. Um, never say never as to whether we'll do a, you know, different or go a different direction with future SKs, but we're pretty comfortable where the decisions we've made this time and in particular around not making, you know, radical changes to to the design because we want to get these things built. And as anyone knows that has built hardware, it's just a long lead time to get things locked in. And, you know, software is very very different, but hardware you got to be, you know, working in in lead times of months and months, right? So, um, that's really the decision-m process behind it. I have spelled it out in the Discord. We've got some healthy discussion, which is great. We welcome that. But, uh, no, we'll have a single speaker in this version of the glasses.
And the camera will be for an AI sensing purpose only. Uh, it won't be a camera that you'd want to use or should use for videos and photos and that sort of thing.
>> By the way, privacy is really matter to us. We want it to be as safe as possible. So, we want to give the users, you know, this experience.
Wear personal glasses device that want to give the user the best experience but only for the user that wear them not for the environment. Supposed to act as regular glasses when you refer in the environment.
>> Okay, a couple more questions. This one I'm just going to kind of combine two of them, but someone asked how the eye tracking and overall control of the UI is. And then someone else asked about if there's any like plans for like control units like a like a ring or a band or anything. Also like the eye tracking is that something you can control everything in the glasses with or is only certain applications or like >> eye tracking has two main uh use cases.
The first one is controlling when we can implement whatever we like in the companion application like swiping slides hoover to select if you look at something and the SDK actually gives you all the information. So we will be more than happy for the community to implement their own you know eye tracking invention. It would be great.
We have several concepts. We're still evaluating some others but definitely controlling the UI will be really really easy. The other main uh use case of the eye tracking is actually have a better context for the AI agent.
Imagine yourself. I'm uh looking at a garden and I'm going to ask the AI take this AI sensor and uh tell me which flower is it and you select 10 different type of flower. So how do you know which one I'm you know looking at with the eye tracker the eye gets all the scene and the exact flower I'm looking at. So we can answer most more precise about my query. It's the same for the dress. I'm looking at you know in store looking at several shirt. So how will they know which shirt I'm exactly looking at? So this is the one of the main feature of the eye tracking in addition to controlling the UI.
>> Okay. So for like the eye tracking as far as controlling the UI, you can do like if you had like a ring or or something, you can do basically the same things like you can look at things, you can select them, you can do all of that just with your eyes.
Um, what about >> line of sight? Yeah, line of sight.
Yeah.
>> Well, so what about like asking AI questions? Is there a way to activate that with eye tracking or with uh like voice commands or anything like that?
Yeah, it's a matter of uh UIX, you know, we have all we have line of sight trigger for example, we can do like this and we activate the UI. So, it's not there's not a technical barrier here.
It's about it's just you know UX decision we have to make. So, we're still evaluating some triggering for activating the AI but you know all the options are on the table. We definitely evaluated the line of sight and eye tracking and some other gadget like you know handling the touch or gazing at a single point. We still evaluating song but all the options are on the table for sure.
>> Just to give you >> by the way >> by external accessories since for example for dedicated application like go or sport. So since the SDK runs on the phone, it's really easy to integrate any Bluetooth accessory because the glasses are connected to the phone, the accessories connected to the same application. So we can receive events for any accessories and just send commands >> to the glasses. So actually the the phone application is like a gateway. It connects the accessories and the glasses. So every bill device can be used. This is how for example we connected the cycling the power meter we received the you know the rad the gin rider exactly in the same way the comment I was going to just throw in was um this is sort of a reflection on several questions we've had um there's a lot of really great questions about hey can you show me you know what the app looks like or can you show me exactly how this works and in a lot of the cases we're still working through different iterations you know, testing. There's like it's called AB testing or ABC testing, different um ways of activating. So, for example, how do you activate the AI in a hands-free speech free way, right? If you got the eye tracking enabled, what's the way to do that that feels comfortable and works the best? And so, uh for those that are on the Discord or in the comments, we're not at all being evasive. We just haven't figured that out exactly yet or we're still working through various iterations. The good news is this is software, right? we can we can do these you know this work right up until a fairly late stage. Um and our goal here is to sort of put the most research we can in and the most thinking and thought as to how it's going to work. Um but yes, for those that are asking and are not getting an immediate qu immediate response, that's the reason why we're still working through it. Um and we'll we'll explain that in the Discord. But um it's kind of exciting.
Like it's exciting seeing what the options are. I mean, you've played with the the swiping and the scrolling and the selecting.
>> Um And so that's one way of doing things. Uh you know there's others we're testing. So uh it's exciting to see what's what's possible.
>> Sweet. Okay. So there's two more left.
The first one is just how high FPS is the display or like the refresh rate >> refresh rate of the >> Yeah. 66. But uh it renders it's the rendering cycle of the display on the glasses. You can update the information on the glasses. uh depend every 15 millisecond or whatever. So you know it's really near real time. Okay. It's remember it's everything through Bluetooth. So actually it's really really efficient here.
>> So you said you said 66.
>> Yes. On the developer portal I think.
>> Yeah. That's like a little higher than like a a TV. TVs are usually like 60 hertz, right? But like I see it's around 60. I can get back to you with the exact number, but it's ceramics.
>> In any event, it's certainly high enough that you don't see flicker. It's not, you know, there's no issue like that.
It's a very uh compared to a TV, for example, it's a high uh refresh rate or higher refresh rate.
>> Um the screen looks good. Yeah.
>> And if you saw the I don't know if you saw the earth rotating earth demo, you don't see any lagging or you know. And then the last question someone asked is if there will be progressive lenses available by release.
>> This is one of the top questions actually. Um it's a great question. Uh and the answer is we are absolutely looking into it. Um progressives are just one level of complexity higher than the other you know flavors of prescription and transition and so forth. Um we absolutely get the desire.
I actually wear progressives myself. So I'm I I would be a customer of it myself. Um, so take it from us that we are chasing this down. Um, we just haven't come back with an answer yet because we don't want to give misleading information. Uh, or if we say yes, the first question we're going to get is pricing and availability and we don't know that yet. So, um, take it from us that we're absolutely working on this. I will also tell you actually one of the advantages of not doing a wave guide, which of course is a very custom lens that requires all the, you know, the optics in the lens. Uh the fact that our lens can be a regular normal everyday um you know substrate optical substrate uh means that the big lens makers in the world are able to make lenses for these glasses. Now right now we have a custom relationship with the lens maker but uh is it possible that we would have you know branded named lenses from the world's largest lens makers uh providing you know custom lenses for the glasses absolutely possible and uh and in some cases in conversation. So um the short answer is we hear you. We get it. Uh we're working on as hard and fast as we can and as soon as we have information we'll share it. Uh what I have promised though is I'll give an update by a day or two before the end of the campaign.
So that for folks that are uh that are trying to make a decision based on this, we'll give them as much as we know at the time. But uh as it stands right now, we're trending in such a way that for example, this is not the answer yet, but for example, it might be yes, we can do it, but we don't know pricing and availability just yet. Or yes, we can do it. Availability will be a month after the, you know, glass is shipped and we don't know pricing. whatever it is, we'll tell people as much as we know at the time uh so they can make the best decision with the data they have.
>> Great targeting.
>> Yeah, likewise. Thanks again. Really appreciate it. Um you've been super helpful for the campaign and you've obviously got a dedicated band of followers, so we love that.
>> You guys have you guys have raised more than a million.
>> It's going gang. So, uh in any event, we'll uh we'll keep pushing. So, sweet.
>> Good stuff. All right, Tim. Thanks, mate.
Yep. Thank you guys.
>> Thank you.
>> So that was the interview. Pretty crazy stuff. I love that you can code for these using things like Swift and Xcode or Android Studio. I've made a ton of apps in Xcode and I'm super familiar with it. And the idea of building apps just by talking to the glasses absolutely crazy. And yeah, all the eyetracking stuff and the thought that's going into like how to activate the AI the most simple way without having to like even say anything. It's all super exciting stuff. And I still can't believe these glasses for the AI Pro version are $389 right now for a full color display. That sounds like a 66Hz refresh rate. Honestly, shocking how cheap this is. So, if you guys want to pick it up, it is down in the description. You can also probably go down there and go to the Kickstarter and and use that to find their Discord if you want to join that and ask more questions about them and be kept up to date on the future, all that stuff. But yeah, these glasses are going crazy. If you want to pick them up, they're in the description. If not, that's totally fine, too. But you're going to want to make that decision really fast because like I said, there's only 3 days left at the time this video is uploading. Thank you so much for watching.
Related Videos
Agentforce NOW AMA: Build with React and Salesforce Multi-Framework
SalesforceDevs
490 views•2026-05-28
How agent o11y differs from traditional o11y — Phil Hetzel, Braintrust
aiDotEngineer
450 views•2026-05-28
WEB TECHNOLOGIES UNIT-2 | Degree 4th sem BCOM Computers web technologies unit-2 full explanation💯✅
LearnwithSahera
1K views•2026-05-29
More tests are always better? How to use AI to identify tests that bring little value
Alliance4Qualification
335 views•2026-05-29
Search Algorithms Explained in 60 Seconds! 🤖💨
samarthtuliofficial
218 views•2026-06-01
People of Game of Thrones using JavaScript DOM
AltCampus
296 views•2026-05-30
Introduction to Problem Solving Part - 1 | Lecture 1 | Intermediate DSA
ascensionix
107 views•2026-05-29
So What's Odin Lang Even Good For
TechOverTea
131 views•2026-06-01











