S4 EP1 Personalizing Student Support: Jose Moreno on Innovative Educational Technology

PODCAST Season 4 EPISODE 1 

Jose Moreno
CEO
Neulight 

Quote of the Podcast: “You miss 100% of the shots you don’t take” - Wayne Gretzky  

Introduction of Guest BIO –  

Jose Moreno graduated from Harvey Mudd College and began his tech career at Microsoft, working on services for over 500 million users. At Netflix, he contributed to features like "skip credits" and led the Account Sharing initiative, adding over 40 million new subscribers. As founder and CEO of Neulight, Jose now focuses on transforming education through technology. His company’s product, IntelliTier, leverages AI to empower educators to provide personalized student support, aiming to make a positive impact in education. 

Interview 

Agents of Change: Leaders/Innovators 

•  30,000 Ft. View – Why do we, as a society invest in education? 
• What drew you to education? 
• - The transformative potential of AI in personalizing education 
  - Applying big tech insights to solve educational challenges 
  - Innovative approaches to student support systems 
  - Balancing technology integration with soft skills development 
• What are the biggest challenges to you? 
• What would you like decision-makers to know? 

Podcast/book shoutouts 

Also, can you use this link for our website:
https://www.neulight.io/?utm_source=educationperspectives&utm_medium=podcast&utm_campaign=podcasttour&utm_id=11 
 

Other relevant links are:
https://www.linkedin.com/in/jose-neulight/ 
 

https://www.linkedin.com/company/Neulight 
 

https://www.facebook.com/profile.php?id=61559694274342 
 

https://www.instagram.com/neulight.io/ 

Transcript

Interviewing Host [00:00:02]:
Welcome to education perspectives. I am your host, Liza Holland. This is a podcast that explores the role of education in our society from a variety of lenses. Education needs to evolve to meet the needs of today and the future. Solving such huge issues requires understanding. Join me as we begin to explore the many perspectives of education.

Liza Holland [00:00:28]:
Jose Moreno graduated from Harvey Mudd College and began his tech career at Microsoft, working on services for over 500,000,000 users. At Netflix, he contributed to features like skip credits, the account sharing initiative adding over 40,000,000 new subscribers. As founder and CEO of New Light, Jose now focuses on transforming education through technology. His company's product, IntelliTear, leverages AI to empower educators to provide personalized student support, aiming to make a positive impact in education. Well, Jose Moreno, welcome to Education Perspectives.

Jose Moreno [00:01:10]:
Thank you for having me. Really excited to be here.

Liza Holland [00:01:13]:
Well, we're really excited to have you and wanna kick you off with the question I ask every guest. From a 30,000 foot view, why do you think that we, as a society, need to invest in education?

Jose Moreno [00:01:25]:
Kids are our future. I mean, 1st and foremost. Right? The thing that excites me the most about humans in general is the amount of innovation that we're capable of. The amount of innovation that we've seen over the last 50 years has been tremendous with the introduction of computers, and I just get more and more excited about what the future is gonna be. And in order to continue to push things forward for the human race, we have to have an educated base. People that understand the fundamentals and the creative that have the creative ideas to continue to push things along. And I don't know what's possible in the future, but I'm excited to see it.

Liza Holland [00:01:59]:
That is great. I agree with you entirely. Entirely. So you have a tremendous background in computer design and that sort of a thing and machine learning, and so you have now decided to kind of apply that to education. What drew you to that space?

Jose Moreno [00:02:19]:
To be honest with you, a little bit of it was strategic. So anytime I've talked to educators historically, they've complained a lot about the type of software that they have to deal with day to day. So for for a long time, I knew that there was some kind of opportunity there. When I started thinking about creating my own company, I started wondering what sector to go into. And, education just kept drawing me in more and more. And so what I ended up doing was I reached out to my cousin who's a vice principal for a school with, kids with special needs. And I just asked her, like, hey, what's your biggest problem day to day? Like, what what's going on? What can we do to help you to make your life a little bit better? And then she started telling me about this new intervention system, multi tiered system of support or PBIS, and how that's a data driven system to help provide additional benefits for students. So when it's done with Fidelity, we see an increase in attendance, grades, graduation rates, which is all great.

Jose Moreno [00:03:18]:
But what she told me was that a lot of people were struggling in actually figuring out how to implement it, how to come up with intervention strategies. And so when she said that, not only did I love like the potential benefits of it, I also immediately started seeing a solution in my mind of how AI can help provide personalized student support if you have a complete product that is able to make data collection simple, centralized in data collection. And I know we're jumping into things a little bit more, but all of this just really excited me when I was speaking with her about it. And and so after that conversation, we just decided dive straight in, talk to more people in the education space, and we just kept getting this feedback more and more. And so it just got us more excited, and so we're fully invested at this point.

Liza Holland [00:04:04]:
Well, it's actually a perfect segue into, you know, you have some pretty amazing solutions to be able to offer. Can you talk a little bit about your system, and why it's a good option in this sea of everybody trying to dive into AI?

Jose Moreno [00:04:21]:
Yeah. So, my last nine and a half years were spent at Netflix, and I love so many things about that company and that product. And one of the biggest things that I saw while I was there was that a number of people just love Netflix. And it would always kind of, like, like, caught my curiosity. Like, why do people love Netflix so much more than, let's say, Amazon Prime Video? Or I don't really wanna, like, talk about competitors, but I mean, there is, like, a bit of a difference there. And it turns out that a big part of it is that it's just so simple to use. It's always there for you when you want it, and it's fast. Right? And it makes your life a little bit easier when you wanna wind down.

Jose Moreno [00:05:00]:
At the end of the day, you just kinda turn it on, press play, and it immediately starts playing. You don't have to really think about things too much. And so we're bringing in a lot of that philosophy into EdTech. We wanna build products that make your lives easier. So we can help you with brainstorming, how to personalize student support. But we also wanna do it in a way where it's a joy to use or something that you don't really think about very much. Right? Data collection. Now that I'm like really diving into the sector, I'm seeing more and more memes from teachers that are like, oh, data collection, data collection, data collection.

Jose Moreno [00:05:33]:
And then nothing ends up happening with it. Right? So for me and for my tech background, what I know and what I've seen is like, yeah, you can collect a lot of data, but that's not necessarily the end goal. Collection of data doesn't really get you anywhere. What you have to then do is take that data that you're collecting and turn it into action. And so we want to we what we see here is an opportunity to make that data collection piece as easy as possible. And then what might be missing today for a lot of different scenarios in education is turning that data into action, into things that actually end up benefiting not only the educators, but but also our students.

Liza Holland [00:06:11]:
So if a teacher was going to from a teacher's perspective, what will their experience be like in this realm?

Jose Moreno [00:06:20]:
Yeah. So we have, Netflix user experience designers. So once again, they focus on making it as simple as possible. But what we've done up to this point is we've made the platform so that it's configurable, so that it adjusts to your school and to your settings. And then and so what that basically means is that we've designed the form in a way where you get a lot of drop downs. You just kind of pick and choose what you need to put in, and then you hit save. It saves, and you're out of there. So you go in, boom, save.

Jose Moreno [00:06:51]:
Get back to focusing our students. You have to worry about organization. You don't have to worry about any like, whether or not it's gonna be there tomorrow. If the nice thing about the drop downs is that you don't have to spend a lot of time typing out what the antecedent. We follow the ABC framework right now. So what the antecedent is, the behavior, the consequence. You just choose, pick, and choose what's relevant, get out of there. And then once you start focusing on strategizing the interventions for students, that's where our AI component can come in and say, you can just ask it like, what interventions do you recommend for the student? And then based off of the MTSS frameworks that it has been trained on, it could just give you some ideas for intervention strategies for that student and how to move things forward.

Jose Moreno [00:07:32]:
As we continue to build out IntelliTear, we wanna get more and more of a whole child view. And so we wanna bring in assessment information, information about the environment, really try to understand what the motivation is for certain behaviors to continue to be able to provide the most effective strategies.

Liza Holland [00:07:52]:
I tell you, I think that a lot of teachers will be really excited to to hear about the availability of this because the more teachers I talk to, the more for some reason, post COVID, student behavior issues are probably one of the biggest challenges that that teachers are facing. So this kind of allows you to customize per student. Right?

Jose Moreno [00:08:14]:
Exactly. It's a per student view and per student recommendations. It's not just like some wide umbrella of recommendations. It actually looks at the student data to figure out, hopefully, what's the most effective way to go forward. And I do see this. I've recently also started watching Abbott Elementary where I think they talk about it as well. But there are a lot of memes on the Internet. I've heard it from, the people in education that I speak to.

Jose Moreno [00:08:37]:
Like, things have changed since COVID, and I find that quite interesting. And so I'm excited to learn more and more about what sort of impact our product can have. But I think more than anything, what I wanna put out there is the way that we're trying to approach this is that it's a partnership with you and all the educators out there. Like, you tell us, give us feedback. What works? What doesn't? Are we doing great job? Are we doing a horrible job? I'm more than happy to hear, like, people screaming at me and say, like, this is giving the worst recommendations in the world. It's just theoretical. It doesn't actually work work in practice. Like, oh, great.

Jose Moreno [00:09:09]:
Okay. Good to know. Let us, like, dig into it. Try to figure out what's going on and adjust to ultimately get to, hopefully, the strategies that do work in practice.

Liza Holland [00:09:20]:
See, I love that. I think that that we have lost our ability to iterate to some extent in education. Everybody's so busy trying to implement all of these different standards and everything else that we don't have time to sit there and go, okay. Well, that didn't work quite the way I wanted it to. What can we do to make it better? And creating those opportunities for students is actually really powerful too. So does the platform actually have a student facing interface, or is it most mostly is that something that might happen in the future sort of thing?

Jose Moreno [00:09:53]:
Yeah. I mean, it's something we could definitely talk about. But right now, we're focused on on the educator side. I think a lot of focus recently with AI has been on the student side. Right? Like personalized student lessons, personalized tutoring, and so forth. But what I'm not really seeing is a lot of focus on the educator side. What can we do to make our teachers' lives easier, our administrator lives easier? And so that's what we're focusing on initially.

Liza Holland [00:10:17]:
And I'm so glad that you are, because really, you are you are hitting some major pain points that I hear on a pretty regular basis as well. I was kind of intrigued that you were talking a little bit as we were setting this interview up about that kind of balancing of technology integration with soft skills development. What are your thoughts and dreams around that one?

Jose Moreno [00:10:41]:
I mean, at the end of the day, I think what I really focus on is objectivity. So looking at what the data says and looking at cause and effects from that data. So and that may or may not, like, fully apply to the soft skills side of the world. But, ultimately, what I believe is that we can record the data to say, yes, this does work or no, this doesn't work. And then we can adjust from there. Right? And so, like, how we go about in terms of strategies for teaching kids soft skills, there are different strategies out there. Now the question is, which strategies work? And that's where I think that the data driven analysis component can really help. It's not so much about us saying, this is a new strategy that works to teach kids soft skills.

Jose Moreno [00:11:30]:
That's not what we're doing. What we're doing is trying to figure out, okay, cool. What are all the different strategies that are out there to help kids with soft skills? And then us trying to figure out what's most effective. And to be honest, it might be a mixture of them. In certain scenarios, it might be this strategy over here. In different scenarios, it might be this other strategy. And so that's where the personalizing intervention strategies based off of that student's data comes into play to really figure out which one of those are most effective.

Liza Holland [00:11:58]:
That makes so much sense, and I love the ability for AI to be able to help to kinda be a thought partner. Yeah. You know? I mean, that that I know I personally use it a lot for, brainstorming and making sure that I've incorporated all of the different, things that a client wants in a in a particular piece, that sort of a thing. So tell me a little bit, what would you say to educators that are flat out afraid of AI? Their instinct is to just ban it from the classroom. What would your advice to them be?

Jose Moreno [00:12:33]:
I think we honestly have to embrace it. Like, the future is here. The the parallels that I've been drawing for the evolution of AI and technology is similar to 2 huge movements that we've seen historically. And you can see the impact that that it's had. The first one is the Internet. And so you ended up having the dotcom boom. And in the late nineties, you had so many companies just building a website. And just the fact that they built a website didn't have to be a useful website, but just building one made their evaluations go out of control.

Jose Moreno [00:13:02]:
And so that's what end up creating the crash in 2,000. Because at the end of the day, you end end up realizing that a lot of companies don't have substance behind it. They just created this little thing. And I think that you're seeing a lot a lot of that with AI today. They'll just put on the chatbot on top of something and say, like, cool. We're an AI company. Like, I don't think that's the right way to go about it. And then the second one is the second movement is smartphones.

Jose Moreno [00:13:25]:
By creating smartphones or the iPhone and bringing apps onto the phone, you just made so much more software, so much more easily accessible and powerful. And so now you can be on Facebook or different websites, all these social networks, games, and so forth on your phone at any given point in time. And so it really move things forward. And I think that AI is along the lines of those two things. Really, what AI is doing, it's democratizing knowledge. You have these models that know a lot about a lot of different subjects. Do they make mistakes? 100%. They make mistakes.

Jose Moreno [00:14:01]:
But so do humans. Right? Anybody you learn from, they also make mistakes. But that that democratizing of knowledge is so powerful. It really makes things so much easier to consume. And not only that, being able to learn different subjects can be personalized to you personally, and ultimately make it more effective for you to be able to understand things. And so I would say don't be scared of AI. It really pushes things forward. The other parallel that I draw is like kind of to a calculator.

Jose Moreno [00:14:31]:
Calculators made it so much easier for people to do math. Right? You just hit in a few buttons and you can all of a sudden multiply 6 digit numbers against 6 digit numbers right in the palm of your hands. But ultimately, the question of like, how do we teach kids to know the fundamentals about math when you have a calculator, when they can just put that in? That was a challenge back in the day when calculators really started becoming available. And so I would say that's a tool that you should really think about. Like, how do we approach calculators and how do we approach it with AI? AI makes it so much easier to write essays, but kids don't need to understand the fundamentals. One of the strategies that I've heard over the last couple of months that I found really interesting was suggesting, maybe tell your kids to go home, write an essay using AI, and then come back into a classroom the next day, and then break it down with them in the classroom. Right? Like really go through it line by line. What would you have changed? What do you like? What don't you like? And then you still start to dive into the fundamentals of writing an essay.

Jose Moreno [00:15:33]:
But at the end of the day, you're leveraging new tools that are there to help everybody kinda move things forward and move a little bit more faster.

Liza Holland [00:15:40]:
I think that is great, and I agree with you. I think that the opportunity that it presents, if you just shift your perspective just a little bit, is really pretty amazing. You obviously are already looking at a solution for an immediate pain point for educators, they have a lot of others that AI might be able to, to solve as well. I was amazed at Cheggpt's ability to create lesson plans and rubrics and all these different types of things in seconds that used to take them hours. But so many educators don't even realize that that is available to them yet.

Jose Moreno [00:16:20]:
Can I ask you something?

Liza Holland [00:16:22]:
Yeah. Please do.

Jose Moreno [00:16:23]:
What do you think is the base of the fear? What are people most scared of in AI?

Liza Holland [00:16:27]:
I think that it is probably very based in the fact that the students are going to know it a whole lot better than this the teachers are from the get go. And that that kinda leads me to that piece of education where we've gotta get away from teachers being the sage on the stage that knows everything to a facilitator and a coach of learning. You know? That that there are things that you can learn from your students. And that's a real paradigm shift for a lot of educators. And I think it's it's the ones that feel like they need to be an expert in everything, and we've gotten to the place where nobody can be an expert in everything anymore. It's the pace of change is just, you know, going almost out of control. We need to move from that old system to kind of how you process information as opposed to the details of that information, because that information might be obsolete next year. You know? And I think that that's that's a lot of that's a lot of it.

Liza Holland [00:17:38]:
And I also think that in this day and age, there's a lot of misinformation and disinformation out there. I'd love to hear your thoughts on how teachers and students can kind of sort through the hype, and, really, what do you need to look for if on a model that might be trustworthy to use?

Jose Moreno [00:18:01]:
Yeah. So the way that I think about AI I'll give you, like, a little bit of maybe a philosophy shift. I think the way that you should think about it, and teachers could, maybe associate with this a bit, is, like, somebody that just came out, just got their masters, and is now TA ing in your classroom. Right? They're so knowledgeable about so many things that maybe you didn't learn while you were in education yourself, but they don't have is the experience of all the years that you've been in your position. But they are knowledgeable, and they are eager, and they are willing to share all of that information with you. It's just about trying to get that information out of them in a way that's actually useful and effective, and that really can only be done by somebody that has the experience. And so, honestly, if you try Chat 2 PT or any of these AI models or our assistant, treat it like it's a TA fresh out of school. Ask it the questions the way that you would ask a TA, and I think you'd be really surprised in how they come out that way.

Jose Moreno [00:19:02]:
But the second part about, like, them being a TA is that sometimes they do provide information that you based off of experience know isn't necessarily true or like what works in practice. And that's really where your expertise or like or that experience comes into play. So you can't just hand everything over to AI. That's not where we're at today because they haven't lived through what you lived through. But what they can do is help you, and then in combination with your experience, you can be the most effective version of yourself in whatever you're trying to do.

Liza Holland [00:19:35]:
Boy, that is probably the most compelling description and recommendation that I have heard so far. I love that. I love that. And, you know, I I think that also I would add to it that teachers need to be cultivating lifelong learners, and therefore, they need to be lifelong learners themselves. And the idea of kind of reverse mentoring should really come into your perspective, you know, that you have lots of experience and amazing things that you can teach these new folks. But they have brand new things that maybe you haven't had a chance to read up on and learn about, that it can be an exchange as opposed to, you know, just a I'm a mentor, and I can tell you what needs to be done sort of thing. I think that we are changing so fast that it's gotta be that way. We've all gotta keep up and learn new things and kinda learn how to learn, unlearn, and relearn.

Jose Moreno [00:20:36]:
Yeah. Exactly. And I would never tell anybody to just take what an AI response, like, says as a 100% fact. Don't do it. Like, at the end of the day, honestly, 99 plus percent of the time, it is gonna be correct. But that 1% at a time, you still have to be on top of it. You have to really make sure that the information that it's providing you is correct. And so always look at any response that it gives you critically.

Jose Moreno [00:21:01]:
It's a lot of times it's a lot more about just brainstorming with you. Like really giving you creative ideas about different ways that you can approach something that you're trying to solve yourself.

Liza Holland [00:21:10]:
My husband actually hates writing, but he is also doing lots of really innovative things in veterinary medicine. So I've been trying to turn him on to ChatGPT and

Jose Moreno [00:21:20]:
Yeah.

Liza Holland [00:21:21]:
And these types of tools because he just doesn't like the actual physical sitting down and writing it. He wants to talk to somebody and just have them do it. But it was funny. He went to try to, to do some things last week, and AI created a reference that didn't exist. It was a theoretical reference of something that, you know, could support his position on something. You know, he kind of said include references, and it made one up. So I I don't think that we humans are in danger of losing our jobs anytime soon because it does take that critical look at it and making sure to to look at at what the references are, folks out there, or anything that it says. And but you can spend your time really making it awesome because AI does the first draft for you.

Jose Moreno [00:22:12]:
Yeah.

Liza Holland [00:22:12]:
You know? Exactly. And first drafts are never great.

Jose Moreno [00:22:15]:
Like, how many times did you have writer block and couldn't just start writing the thing that you're trying to put out there? You can just ask you to write something, and at the very least, you have, like, a first version of it, and then you see what you like or what you don't like. I have 2 interesting, like, hot off the press kind of tips for for that situation. You can actually say, like, tell it, don't make anything up. Only provide me with, like, real facts as a part of your prompt, and the amount of hallucination that the AI model does drops significantly. Awesome. And it's crazy. But it's like just something so simple, like, don't make it up. Tell me just hard facts.

Jose Moreno [00:22:53]:
If you don't know, just say you don't know. It's fine. Just kinda giving it that

Liza Holland [00:22:57]:
Give it permission. Yeah.

Jose Moreno [00:22:59]:
That confidence to be able to to say I don't know is it does wonders. The second part is ChatGPT last week just announced that they are going into the search world. And so what this basically means is that when you, in the future, ask ChatGPT for something and you ask it to provide references the way that your husband did, it'll be able to go out, do search, search the Internet, find actual references, and then on its responses for each of those citations, it'll be able to link exactly what that citation is to provide more more confidence that the information that it is it is providing to you is real information. And so that's a new version that I think that it's in beta right now, so not fully available. But I'm guessing within a month or so, everybody's gonna have access to that version of ChatGPT. And the technology for this AI movement, it's moving so quick. Like, we understand what the problems are, but anytime you release something new, there are gonna be issues. There's gonna be things that aren't necessarily perfect.

Jose Moreno [00:23:59]:
But as long as you keep iterating and improving and continue to push things forward, what we have in 3 years compared to what we have now, it's gonna be crazy. It's gonna be ridiculous.

Liza Holland [00:24:10]:
Well, and it's already improved so much from the very first time that I tried it to now. It just seems like it's constantly getting better and better and better, and that's just really exciting. A lot of people I talk to don't really understand the idea of prompt creation. Can you talk a little bit about what you think that the models are looking for from, to get the best types of results?

Jose Moreno [00:24:34]:
Yeah. I think some of the difficulty that I may not have faced initially when using ChatJPT versus the general public is that I see a lot of it the way that I see software programming, which is kind of, just to give a quick little background for, computer programming. When you're programming something, you have to tell it exactly what to do, like, line by line by line by line. I actually remember, in my 9th grade English English class, my teacher had an assignment where she said, write me, like, exact directions of how to make a peanut butter and jelly sandwich. Right? And so you would say, like, grab the peanut butter, put a knife in, take the peanut butter out, put it on the bread. But just with those steps to there right there, you didn't say, like, where do I grab the peanut butter from? Right? So there might not be anything on the table, and you say, grab the peanut butter, and then your teacher tries to grab the peanut butter. There's nothing to grab. Boom.

Jose Moreno [00:25:33]:
You failed. And so in computer programming, you have to be so exact to say, like, wake up and get out of bed, but I like your clothes. Go to the kitchen or the pantry, open up the cabinet, grab the peanut butter out of the bottom shelf of the cabinet, walk over to to your table, open up the peanut butter, go to your knife drawer, grab a knife, go back to the table, stick the knife in. And so you basically have to be, like, super detailed in terms of what you actually want. That's how I think about engineering or, like, the prompts that are created for AI. Don't just assume that the model knows what you want. Like, be very, very exact about what you're asking for it from it. So even something as simple as, like, don't lie to me or, like, don't make anything up.

Jose Moreno [00:26:21]:
Right? That's something that we all take for granted when we're having a conversation. We just assume that the other person is not gonna lie to us. But a model itself, it doesn't know that. It doesn't know that that's not one of the rules that that you're, like, hoping that it doesn't do. So if you tell it exactly that, write as many details as possible. Like this is the background. This is who I am. This is the voice that I usually write my emails in.

Jose Moreno [00:26:46]:
This is the subject that we're talking about. This is who I'm reaching out to. This is their background. Right? Like more and more and more information. The more information that you provided, the more it understands what you're actually trying to achieve. Anything that you don't give it information about, it just assumes and makes, like, best case assumption for what it thinks you might want. And then that's where the disconnect ends up coming in between the responses that you're getting that you may not be happy with and one that it can absolutely be capable of that you would be happy with.

Liza Holland [00:27:17]:
You know, my favorite trick, as I have been on my own learning journey, is to, you know, you write the whole prompt, and you think you've got it, and then it's like, well, no. What they're giving me doesn't include this piece. Well, I never included that piece. So rewrite to include x y z.

Jose Moreno [00:27:33]:
Exactly.

Liza Holland [00:27:34]:
I use it all the time, and it does. It makes it better and better and better each time, and it allows it to iterate in seconds where, you know, you may have to completely rewrite something or or not use it at all. And I just find it a fairly magical experience.

Jose Moreno [00:27:50]:
And that's the second part of it. Right? Like, whatever you end up getting the first time, it's gonna be a first draft every time. And so what do you do with your first drafts? You read through them. You look at them. You analyze them. What do you like? What don't you like? And then you change things, and so you just continue to tell it how you wanted to change things to ultimately get to, like, what you're hoping to to arrive at.

Liza Holland [00:28:08]:
Excellent. Well, tell me as getting kinda back to you and your initiative and and everything, obviously, you had a really good thought partner in the, the the prince vice principal, your cousin?

Jose Moreno [00:28:20]:
Vice principal.

Liza Holland [00:28:20]:
Vice principal.

Jose Moreno [00:28:21]:
Acting principal this year, I believe.

Liza Holland [00:28:23]:
Oh, hey. Congratulations to her. But education can be a a little bit of a morass to navigate as far as getting it out to to everybody and whatnot. What have the biggest challenges been to you in starting this new company and and getting it out into the world the way you'd like it?

Jose Moreno [00:28:41]:
Yeah. I think one of our biggest challenge so far has been not taking things for granted. So or taking things for granted. Along similar lines where I come from a technology background, and so I look at what we built and, like, yeah, super intuitive. You can do this, that, and that. Sometimes people look at it and it has like a for example, our initial version had a blank page on our assistant. It's like, I knew that you can ask it more information about MTSS frameworks, or you can ask it to brainstorm with you about different tier 2 intervention strategies that you could put into your school. Very powerful thing.

Jose Moreno [00:29:16]:
Very useful. But when somebody most people look at an assistant like this, they don't know where to go. Right? And it's like, okay, that's nice to have. And so we saw that and we started, building in some prompt suggestions into the system. And so, now when you look at it, you're gonna see different conversation starters for what you can potentially do and then start diving into that. And then as you're having the conversation, we're also gonna include some suggestions of how you can move the conversation forward or how you can pivot into, like, a different type of conversation to once again, keep those creative juices flowing to try to maximize the value that you can get out of the assistant. Just kind of along those lines, it's continuing to hear the feedback that we're getting from our initial customers, looking at what they're doing and then figuring out how can we make this simpler and simpler and simpler to use to be able to get the value out of it. Because what our hope is is that if we make it extremely simple and extremely effective through data driven concepts, then you can provide really effective personalized student support for your students in a really quick timeline that should take some of the load off of our educators and allow them to focus on our students, which at the end of the day is what everybody wants.

Jose Moreno [00:30:32]:
Right? We all want you to be focusing on the students and helping them learn and and move them forward. And so, in any way that we can make things easier, we will continue to push to do that.

Liza Holland [00:30:42]:
So tell me your thoughts about student privacy. You mentioned that real quickly, but I'd love to hear a little bit more.

Jose Moreno [00:30:49]:
Yeah. And like I mentioned, there are a lot of companies out there that are just kinda throwing in a chat GPT integration with, like, a chat on top of whatever they already had and then calling themselves an AI company. We're not seeing it that way. In order for us to truly combine the technology with the education side of the world, the AI has to be deeply integrated into our platform. And, in order to deeply integrate this super sensitive data into our platform, we need to make sure that we're taking care of this data. So we've built a FERPA compliant platform that is hosted on Microsoft Azure servers all within the United States to make sure that doesn't go anywhere. We we actually do not use ChatGPT. We don't use, or at least OpenAI API Servers.

Jose Moreno [00:31:37]:
We have a model that we host within our Microsoft Azure Servers that then our back end servers, and maybe is a little technical, end up talking to. But that data is never used to train a model. It never goes outside of our servers. We don't build anything with third party vendors. Everything is built in house. And so putting all of that together really allows us to make sure that we're taking care of that data because this is super, super important data, to make sure that we handle with care.

Liza Holland [00:32:08]:
Sounds like a really good thing for decision makers to be asking about as they move forward because that that is a seriously important piece of the puzzle.

Jose Moreno [00:32:17]:
A 100%. I I read a a recent sort of leakage that ended up happening with an AI that was basically making into servers outside of the United States, and that was a big no no, right, in terms of, like, this super sensitive data. So, yes, anybody that is providing you with, like, an AI, type of solution, ask them, how do you power this? Like, where where does the data live? How do you what do you call in order to get the responses from this? If they don't have AI experts within the company so my cofounder has a PhD in machine learning. Like, it was a very big priority for me as we were getting the company off the ground to make sure that we had somebody with that expertise come in. But if they don't have AI experts within the company, then a lot of times they use 3rd party vendors for that. And once they start using a 3rd party vendor like OpenAI to power their chat JPT experiences or their, AI experiences, it becomes a bit of a concern in terms of what's happening with that data. And also a big reason for why you shouldn't be asking chat GPT for interventions that are strategies. Can it do it? Absolutely.

Jose Moreno [00:33:20]:
But in order to do it effectively, once again, you have to give it all the information in the world. And once you start doing that, you're kind of offloading some very sensitive data into, a service that you may not be able to trust.

Liza Holland [00:33:34]:
Good point. Thank you so much. That is fabulous. And, again, that iteration, I love it. I love it. And it's hard because you see that on software development a lot, is that the what the developers think is intuitive is not necessarily what the end user finds as intuitive. So it's it's great that you have a foundation of regular feedback like that.

Jose Moreno [00:33:59]:
Yeah. Luckily, that's something that I've always had a bit of an advantage against my coworkers in the computer space, computer science industry because my parents weren't technical. I had to bus to my middle and high school for 20 miles in the middle of LA traffic every day. Like, I experienced, I guess, what most people experience day to day kind of growing up. And my technical skills came in more like late high school and definitely in college and then, of course, my career. And so that combination of both of those experiences has turned out to be quite an advantage because I understand what non technical people go through day to day, but I also understand how to potentially solve some of those problems on the technical side of the world. And so that's always been an advantage. And now I'm hoping that it's an advantage that our company has, period, versus other, EdTech competitors.

Liza Holland [00:34:49]:
Well, you know, building that bridge is what it's all about. And sometimes, you know, even the languages are completely different. Being able to figure out what, you know, all those acronyms that you threw out with PBIS and supports and all this other kind of stuff, I'm sure that your computer folks knew nothing about that when they first started. So it's making sure that that you even define terms similarly or can understand what you're saying versus what they're saying. It could be a real challenge. But it sounds like you've got a super effective, you know, feedback pattern to be able to continue to do that, which is amazing. And I know that educators will definitely respect that because so often the software systems that they're presented with are you have to figure out how to use it as opposed to how can we make it better to be able to better suit you. Exactly.

Jose Moreno [00:35:43]:
That's crazy

Jose Moreno [00:35:43]:
to me.

Liza Holland [00:35:44]:
I think that's a little magical. I know. It is. But it happens. It happens. I'm sure you've heard heard of lots of examples of it. But, boy, this has been a fabulous conversation, and I thank you so much for looking for how you can be a solutionary for the education space because it is it is sorely needed. The teachers out there are so overburdened with so many things, and this sounds like a really amazing tool, to make it just a little easier for them.

Jose Moreno [00:36:14]:
Thank you. Yeah. And if anybody is interested to, like, really work with us, especially initially while we continue to build early on, we're extremely excited to talk to anybody and everybody that wants to. So please reach out to me. I'm on LinkedIn. I think it's one of the best places to contact me. So Jose Moreno, and then our company name is Neulight, n eulight. You can also go to our website, which is neulight.io.

Jose Moreno [00:36:40]:
And we have a contact us form there that you can send out a message, and I'm looking at every single one of those as well. And then finally, we do have a free version of our product available on our website. It has almost everything that we talked about today. What it doesn't have is the ability to have multiple people sign in to the same school. So, basically, one person can put in all the data for a specific student, speak with the assistant, get personalized intervention recommendations. But the counselor or other teachers that that student also sees won't be able to see that same data. That's where we move over to to our paid version.

Liza Holland [00:37:17]:
Boy, that's really exciting. I'll make sure to drop all those links in the show notes as well for listeners so you can get directly there. But I love that you are willing to share and let people try it out and, and and see what it can do first. That's really amazing.

Jose Moreno [00:37:32]:
We believe in our product, and we believe in what we're doing. And if we continue to execute well, then I only see us being successful from here.

Liza Holland [00:37:39]:
Well, I can definitely agree with you. And I'm gonna leave you with a final question because this integration of technology is still a little new to our old systems of education. So when you're thinking about applying technologies in different areas like this, especially the ones like yours that actually solve problems that are existing, what would you like decision makers to know?

Jose Moreno [00:38:08]:
1, we're here to work with you as opposed to just giving you some product and saying take it or leave it. 2, your feedback is extremely important to us. And then 3, I'd really love to learn what other problems you have, Not necessarily for, let's say, the next year, but IntelliTear is our flagship product that we're introducing as kind of like an introduction of what new light is capable of. Like, hey, this is new EdTech. This is a bit different from anything that you've dealt with. If you don't believe so, please tell me. But it should be different from anything else that you've dealt with. What's your next problem that we can help solve? And then and really it tells here we'll really show off what we're capable of.

Jose Moreno [00:38:50]:
And then once we start introducing that next product, hopefully, the conversations are a little bit easier to provide people with confidence for what you're gonna get out of our company.

Liza Holland [00:38:58]:
Well, that is a fantastic invitation and perspective for decision makers. So thank you so much. And I really appreciate you taking the time to, to share with us and our listeners here at Education Perspectives.

Jose Moreno [00:39:11]:
Thank you for having me. This was a great conversation.

Liza Holland [00:39:14]:
Thank you so much for listening to this episode of Education Perspectives.

Liza Holland [00:39:18]:
Feel free to share your thoughts on our Facebook page.

Liza Holland [00:39:21]:
Let us know which education perspectives you would like to hear or share. Please subscribe and share with your friends.