Adam Geller on Improving Your Teaching Practice with AI

Hello, you’re listening to the Getting Smart podcast. I am Nate McLennan, and here at Getting Smart, we love thinking about and supporting our partners and all of our friends in education, thinking about the future of learning and innovations for equity. I am really excited to have Adam Geller on the pod with us today.

So welcome, Adam. Thanks so much for having me. Yeah, you bet. And Adam is the CEO and founder of Eddina. It’s a video coaching platform.

They have a library tool and they’re building an AI coach, which we’re really excited about. Adam’s been an educational technologist for a long time. He was a science teacher, author, did some teaching for America work and started his career and doing some news producing in that area.

So, Adam, I always love asking our folks that are on the pod, just thinking about your favorite learning moments. What would you say for that? Yeah, I mean, I think for me, I’ll lean on some science teaching moments that I have experienced.

And I remember this one time we were at the thing was called the World Science Fair or something like this. It was traveling around. It was when I was a teacher in St. Louis. It was a field trip for our students.

So we’re at the Science Museum and there is this scientist who is in explaining string theory, the very complicated physics topics that everyone uses in jokes at cocktail parties that that’s too complicated to understand. And I remember this scientist broke it down in such a way that myself, as well as the students, could understand it.

And I don’t exactly remember what the connection was, but he was literally playing strings on a guitar somehow to explain string theory. So it just was a good reminder of, you know, the best teachers can bring the cold hard facts of science to life for warm thinking, feeling humans and students that we are every day as we go through the world.

And it’s funny what we remember and what’s sticky and it may just be an experience. So I remember doing cat dissection in high school in the basement of my house with a bunch of classmates and I don’t remember anything about cat anatomy, but I do remember cat dissection. So that might be a similar thing to your guitar string theory piece.

Well, perfect. I am a huge fan of Ed Thena. I have a funny connection in that after you started Ed Thena, I and a couple of folks out here in Wyoming started this company called Improved Class, which we have since shut down, but it was about crowdsourcing using teacher video. So I remember distinctly doing a market analysis and Ed Thena was the major player

in the space. And so I had that your logo on my slide and always looked to you for modeling. So super excited to see that you’re you’re moving ahead and helping teachers all around the world. We’d love to know your aha moment for Ed Thena. You had you had been doing some teaching and all sorts of interesting things. And at some point you must have had an aha moment.

So what’s the aha moment for you? Yeah, so I mean, I think it’s some things that lead up to the aha moments. I need to give you that background. I mean, as a first year science teacher, like many first year teachers, you want nothing more than to be the best teacher you can be to help lead your students

to the greatest gains possible. And here I was a first year teacher teaching science in a school where, you know, there was no one with science expertise who could come to my room and actually tell me, how was I doing at teaching the content? And, you know, I always like to credit my principal. We had a great relationship.

It was it was not that she was not coming to support me. It was just that she didn’t have science background, science teaching background. You know, in reflecting on it, I did know a science educator at the Science Museum. But again, he couldn’t come to my classroom to actually help me understand. Did I effectively execute the lessons that he helped me prepare?

So, you know, kind of fast forward. I’m on the National Strategy Team of Teach for America and looking across different regions and at the time thinking through challenges around scaling up access to coaching support that, you know, ultimately you wouldn’t have a chemistry teacher for a chemistry coach for a chemistry teacher in, you know, a rural region.

And how would you solve that problem? Now set the stage. This is back in the time of flip video cameras. And these are like the exciting thing and the idea that you might be able to get a video online, portable video cameras. You know, now we take it for granted on our cell phones, you know,

that was not the case. We forget. And so that was kind of the context. And then I get a link to a viral blog post or something like this that says, listen to a Justin Bieber song, slow down 800X.

So this is not really about Justin Bieber. It’s what I found inside that article, which was an embed of the music player SoundCloud, which allows users to comment on music at moments in time. And I remember hovering over one of the little markers and it sounds something like said something like, you know, it sounds like violence coming in in harmony here.

And even as I say that, you probably have an image, if you will, an audio image in your head of what that might be. And I had this aha moment where I was like, wait a second, I’m not listening to the audio, but I do have a clear picture of what’s happening here. Let me listen to the audio.

Let me line up this comment with the evidence. Yeah, that really is powerful. What if you could translate this set of tools to video and give it to coaches and teachers so that when it’s not possible for that right person to be in the right place at the right time, you still can provide support to teachers in any classroom.

So, you know, it basically helps solve that problem of right person, right place, right time that schools, school districts, teacher education programs all work around right now. Because we know the best way to help a teacher get better is through feedback on their teaching. Right. OK, so I’m still thinking about the flip camera. And I do remember those and they had their short sort of life in the technology market.

So now flash forward to today. That was what is it, a decade old? Is that how old? Yeah, so it sounds like you have about three different distinct parts of it. Athena, can you quickly describe those just for our audience so they get a context?

Yeah, so the I guess the Athena ecosystem, if you will, includes the video coaching platform. This is what a lot of people know us for. These are the tools for commenting on video, collaborating, setting up structured learning experiences. Then we have an organization library product that helps schools and districts curate and then share best practice videos across their organization and securely give

access to everyone in their organization. And then we have a new product that we announced just this spring, which is called AI Coach by Athena. And, you know, this is something we’re really excited about because it really shifts how we’re using the tools for video reflection and video analysis and, you know, helps teachers self reflect on their own so that when there isn’t that coach available to come or maybe no coach

at all, they still can have structured guided support in continuing their professional learning. Got it. We’re going to come back to the AI coach because we’re like I said before, super interested in that concept and think it could have a lot of applicability across the field. So as you think about the last decade and your evolution from where you started to where you are now, what’s been the big learning or any pivots that you’ve made along the way that have been

helpful to get you to where you are now? Sure. You know, I think when folks ask about what is the challenge that we face, you know, we we don’t face a technology challenge. We’ve built technology that is reliable. We can back that, claim up with data. It’s easy to use. It delivers value to the teachers when they use it. They say they like it, you know, all those types of things. So it’s not it’s not a technical challenge. And so

then you have to ask, Well, what is it? And what we’ve come to realize is that our our biggest barrier to going faster with our tools that we’ve had in the past is helping organizations do the change management that’s necessary to change a process related to professional learning. It’s not that the idea of using video is so foreign anymore, but the the behaviors, the structures, the patterns of of doing the work of schools and districts are routines, right? And it’s challenging to

help people basically identify which routines need to change. And so in that way, I think the status quo and learning how to help people shift away from that, you know, helping them really be honest with themselves about what challenge are they working to solve for their teachers in their organization? And what is the implication if they don’t actually make headway on solving that? And when it comes to teacher professional learning, you know, I certainly believe and a lot of educators

believe that teachers in in the learning environments are the biggest lever for enabling student success. So having the highest quality research backed ways of driving professional learning, if you believe that that’s important for your organization, then it might require changing the way the business is done within the organization, because, you know, those sit and get the right feedback. And so I think that the workshops are really easy to plan. But they aren’t

very good for teacher learning or student outcomes. Yeah, yeah, interesting. Important reflection. When you think about so I was a school principal for many years, I did a lot of observations and I had some pivot in my career around starting to look more at the idea of a better way. I wonder if if you’ve seen the use of the Athena in terms of the video platform, are teachers more looking at themselves as the the sage up top saying sage on

stage saying like, look what I can do differently, which may be import, which is important, I think, or are they also noticing students in that interaction sort of love to hear your thoughts on that? Well, actually, maybe what I’ll do is I’ll answer this question with also saying maybe I should answer because I’m as soon as you started saying that, I actually recognized in myself what has been my biggest pivot is actually learning to talk about the power of video as a tool to not just see

teachers teaching but to see students learning. I constantly default into that mode because I have a routine about talking about Athena and how it’s used. But the reality is, is just like you’re asking, teachers are doing a lot of reflection related to understanding student learning, eliciting student thinking, observing discussion with students both between the students and the teacher but also among students in groups. And video provides a really, really powerful way

for teachers to examine what’s happening related to student thinking. Sure, a video camera in the classroom space and the learning space is more than no camera in the classroom. But students today are very used to cameras, right? So after the first 30 seconds, they don’t care and they’re back to doing whatever they need to do. And that teacher can, you know, really revisit those moments of students collaborating with each other and have a way of understanding what type of student learning

is happening that they never had before. Yeah, it seems that as both you and I have talked before and we tend to be myopic when we’re teaching, it’s hard to see everything at once because we’re focused on a particular student, particular topic, particular group of students. So the ability to have a video of that is so amazing to see what’s happening in the places you’re not looking. Is there learning happen when you’re not paying attention to that group? And that’s impossible to

tell if you are not using video. So I see that as a huge advantage. When you think about improvements, let’s pivot a little bit to just humans in general. And how do you connect what’s happening with video based observation to sort of human improvement theory, just the idea of how do we get better at something? Have you connected that in? Or do you have any personal thoughts on that? Yeah, I mean, I think one of the many benefits that video offers, someone who’s watching themselves

is that it allows them to quite literally see themselves in the act of doing something. You know, we somehow never seem to question the idea that an athlete would review game tape footage on a, you know, frame by frame but aspect these days, or, you know, look at a golf swing and actually draw the angles of the club and how the ball is, the club is connecting with the ball, right? You know, we take that for granted, but somehow in the teaching context, we struggle with realizing the

benefit that the teacher who is executing a very complex activity would get from being able to just see themselves. You know, in the moment you are hindered by the cognitive load of the execution of teaching, you are not able to see everything. You may have a skewed perspective on how things sound, how things are working, what the environment actually reads as if you’re someone who’s observing it. And so, you know, what’s powerful about video when you talk about using

video inside learning environments, inside classrooms, is that, you know, video is not the perfect observer, but the video doesn’t have an opinion. It is a mirror on what is happening and it reflects back to the teacher what is happening. And sometimes that can illuminate a lot of great things that are happening. Sometimes that can provide a chance to see some things that are an opportunity for change. But I think at its core, using video as a tool to understand the actual

act of the teaching and the learning that’s happening inside the classroom space, that’s the powerful thing here. Yeah, it makes me think a lot about this ownership piece and autonomy. So, you’re probably familiar with Daniel Pink’s work and he talks about autonomy, mastery and purpose as motivation pieces. And when a teacher is, so when you have another teacher observing a teacher, certainly there’s inevitably bias that comes into that. Even if

it’s a great observation, there’s inevitably bias. But I love the idea that video does not have an opinion. And that then entails or allows for those, the teacher observing themselves, they have autonomy because they can control what they see, what they don’t see and how they’re reacting, they’re getting better around mastery and they see a clear purpose, like they’re trying to be better at their craft so that their students can have better outcomes. So, I sense that there’s

some motivation theory that’s grounded in why this is successful, why we see improvement. Which then allows me to think about how are teachers who have routinely used this over the years, so who are your best users of the Athena platform, how are they measuring growth? Is there a way, is it around self-reflection in the platform? Is it a way to quantify it or is it more qualitative? What are your thoughts on how they’re measuring growth over time?

Our platform has tools to measure growth in qualitative ways and quantitative ways. I think it just depends on how success is being measured. I think one of the special things about Athena and the way that we have four comment types in the video coaching platform, the four comment types, question, suggestion, strength and note, is that even if you don’t have any sort of rating or quantitative understanding of growth, you can still look back and say, wow,

at the beginning of the year, I had a lot of questions related to checks for understanding. And at the end of the year, I have a lot of strengths related to checks for understanding. And so giving teachers the tools and the ability to understand that about themselves is, I think, really powerful. Your question was about what are the things that we have heard? And there’s one particular experience as related to me by a teacher that really sticks out for me.

And this is a math teacher named Taylor. And he was reflecting on his first year of teaching and talking about where he felt he had ended up and all these things. And what he said really stuck with me, which is he was looking at himself at the end of that first year on video and comparing to where he was and seeing all these things that he could still improve as an educator. But that he was also able to understand how much he had improved and that the video had given him

this tool as a first year teacher who, like many first year teachers, feels like there’s a really long list of things I need to do in order to be a great teacher. He had that insight about himself that he was growing. He was changing. He had already grown, should I say. And I think that’s really important, you know, I think is the right word. It’s really important because there’s a lot about being a teacher which can make teachers feel like they may not be

achieving all the things that they thought they were going to achieve. You know, it’s a complicated context right now to be a teacher. And I think to have some objective evidence that you can look at yourself is really valuable because a lot of teachers are doing a lot of great work. And when they can see that about themselves, that’s the type of information that keeps them, you know, recharged and ready to do the very important but very hard

work of teaching. That is so important. I think that I’ve coached a lot of first year teachers who struggle. Most first year teachers, I know I struggle as the first year teacher. And you often don’t feel like you’re making any progress. But the idea that you can have a video to document probably ties into another place of motivation theory, which is around self-efficacy. Like, they believe they can get better because there’s evidence right in front of them that from day one

or the first month or school or two months when they video compared to the end in the spring, there are clear things that have changed, which as teachers in your first year, second year, third year, you tend to forget those because everything’s such a blur in those and at the pace of those. So I really appreciate that story there. I think I’ll add one thing there that, you know, the video also allows teachers to see this about themselves. It reduces their dependency

on having someone external to them point this out to them, right? It doesn’t depend on the school leader or the coach having made these observations and convincing the teacher of what has changed. It empowers the teacher to see this about themselves and develop a stronger, you know, definition of themselves as a continuous learner who is having success in their job. Yeah, sometimes I think that I’ve visited, you know, hundreds and hundreds of classrooms and I

think that over time we get an image of sort of things that we’re looking for and again, acknowledging bias here. But trying to get teachers to that place faster through video would be an important piece because that self-reflection and ownership matters. So right on there for sure. So let’s talk about this, the new platform, the AI coach, clearly it creates an efficiency in the system because then you don’t have to rely on external

other peers or mentors or supervisors. Tell us a little bit about where you are in the development of this and what’s this thing going to look like in the platform? Yeah, so we’ve been hard at work trying to bring something that sounds like science fiction into reality in AI coach for teachers. And we’ve been doing that for more than a year now. And so we made the announcement earlier this spring in order to let people know, hey, we are in the final steps of the development, we’re doing

the testing and this is something that’s going to be ready for teachers all across the country and across the world to have access to starting next academic year. And I think we see the AI coach platform as a tool that basically strengthens teachers as professional collaborators within their school contexts. So yes, it may be that you use the AI coach in between visits from your instructional coach, it could be that you use it because there is no instructional coach.

But at the end of the process, you have a deeper understanding of where you are as an educator and are better prepared to collaborate with your colleagues about what’s happening inside your classroom in order to have the kind of shared and collective efficacy type conversations that are so important within school communities. So the platform essentially, by the way, let me first say what it’s not because there’s a lot of people that are like, oh my goodness, and then crazy thing fill in

the blank. This is not a computer that is going to watch your video for you and tell you what you need to change about your classroom. That is not where we are, that may not even be possible from a technical perspective, but it’s just not what we’re building. What we are building is a tool that coaches teachers to coach themselves. It coaches teachers to self-reflect, to analyze their teaching, to kind of focus in and find one thing to work on, set a concrete next step,

and a plan for follow through. And that is kind of what we hope all teachers will do, of course. And the reality is that that process is hard to do on your own. This is why there are instructional coaches. This is why there are school leaders who are instructional leaders. But there’s just limited capacity out there to help teachers do this reflective work. And that’s where we think the AI coach can support teachers in their continuous learning. So in the system, I’m

suspecting or I expect that there’s prompts that the AI coach provides. The teacher responds. Does that then drive a variety of different responses? Is it reading or understanding the text of the teacher to then provide the next output? Is it a multi-branched type experience? How customized does it? Yeah. Yeah. So right now, so let me maybe describe what the teacher would experience. So I would upload a video. Surprise, surprise. We’re still focused on using

video and analyzing your teaching, getting a clear picture of what’s happening in your classroom. Okay. So a teacher uploads a video. They have a video inside the AI coach platform. Now, maybe you’ve never analyzed a video before. So the AI coach actually just like our instructional coach, hey, what do you want to be looking for when you observe yourself and gets the teacher to focus in? And so let’s say the teacher says, well, I’m really interested in observing myself

for checks for understanding and use of academic language in the content area. Then the platform will make sure they know how to actually do the analysis, not just the technical skill of the platform, but thinking like, hey, you need to go look for strengths and you need to look for opportunities for improvement. Then as they get into the analysis, that’s where you’re going to start to see some of that customization. So right alongside the video, the platform is able to,

based on what the teacher indicates as their professional priority, show customized observation tips for check for understanding and use of academic language. Then they continue through the process. They’re going to be asked probing questions about the comments that they live. They’re going to be asked to summarize those things. And when it comes time to setting a goal, you know, the virtual coach doesn’t say, oh, you know, well, what’s your goal? You know, that wouldn’t be very

supportive, right? So the virtual coach actually breaks that down. You know, what are you looking to change? How are you going to notice that change with students? How would you measure it? Why is this important for this to be true? Right? We just, in the same ways that an in-person coach might drive that teacher’s thinking and probe them, for lack of a better word, to deepen their thinking. This platform is designed to do the same. There are absolutely moments where, based on what the

teacher is working on, like setting a strategy and let’s say the teacher’s like, oh, I’m kind of stuck. I need some ideas. You know, there’s curated content that the platform can provide. So it’s responsive to what teachers want to be learning and constantly adjusting based on, you know, how they say they want the experience to be personalized. And the only other thing I would add here, because I may not have said it explicitly before, is you’re having a kind of

conversational text back and forth with this virtual coach. And I think that’s one of the big things that is different here is, you know, this isn’t a Google form that you fill out, right? This is actually a persona inside this software that is your helpful guide on the side and talking to you along the way. And if you’re a teacher who wants to quite literally talk back, you know, you can click the microphone button and it will do voice dictation and listen to what you’re saying.

That’s amazing. All right. So two questions. So in the AI world, there’s a couple, there’s two things that I’m wondering about. So the first question is around equity and how are you building the bot or the AI system in the back end to make sure that equity is a top priority? And then the second question is, does it learn over time? Meaning, is the data that the teachers are putting in aggregated somewhere so you can make a better AI in the future? Anonymously with

security and all those kind of things, but are you learning, is it learning from the data that’s being entered, which is another feature of AI? Sure. Yeah. So I mean, AI is a spectrum, right? So I think on one end, there’s the fully kind of, let’s call them independent actor type AIs that have been trained on massive datasets. And I think that’s what you were getting at related to is the is the way that the computer program, for lack of a better way of saying it,

trained on data that is going to respond in a way that’s equitable to teachers. So, the type of responses we’re giving to teachers right now are not at that kind of fully independent, unmonitored end of the spectrum. It is really in the realm of using expert and instructional coaches to prepare the computer to respond to teachers’ needs. So, a variety of different resources that represent a variety of viewpoints, the questions, making sure that they’re asked,

shopping those questions around, showing them to lots of different expert researchers to make sure that the way that it’s probing teachers doesn’t bring in any specific disposition accidentally. So, I think that’s how we’ve kind of thought about that. There are some kind of summarization tools and techniques that we will be leaning on, but those will be using generalized models for those kind of taking inputs from teachers and recombining them and summarizing them. We’re not

generating new knowledge for the teacher, right? I want to be very clear, we’re not telling the teacher what to do here. But based on teacher responses, based on what teachers say is helpful, when they say they’re getting the help they need, when they say, oh, you know, that resource wasn’t helpful, but I found this other resource, we absolutely are able to take that as input, use it to strengthen the set of advice and resources that are available to those teachers so that,

you know, they will be reviewed again, but then they’ll expand, you know, the types of things we’re able to offer. Got it. So, you’ve presumably piloted this with a few teachers or classrooms or schools. What’s been the reception so far? We’re in the process with the version right now, the fully independent version, but absolutely along the way, you know, we were asking school leaders, we were asking teachers. You know, I think

one of the funny responses I remember from one of the kind of testing sessions we did with a teacher early in the design process, and the teacher was just like, oh, this sounds like it’s really awesome. Do I get to see the whole thing today? You know, like their response was like, show me more, show me more. You know, and I think the reason, you know, this is me projecting, right? But I imagine that our intent of the design was resonating with the teacher, right? Here you have

a very modern experience in terms of chat style, conversational interfaces, you know, high design. So, it looks like the best type of tools you’re seeing anywhere on the internet, right? And this is something that’s being shown to you as a tool to help you, teacher, like take control of your professional learning and make headway on where you are. Absolutely. You know, that was like, that was resonating with them, right? And so, you know, I think,

I don’t want to be overconfident because there’s a lot of questions we still have to answer, right? But I do get routinely very excited because it’s like the beginning days for us, you know, like back when we made the first version of Athena, the video coaching platform, we have something completely game-changing to offer here. And when educators have an opportunity to see what we’re working on, they say, yes, and can I see more? Because I think there’s a real

desire among all educators to be the best versions of themselves they can be. No educator gets into teaching trying to be lackluster or trying to do the wrong thing for kids. And so, if you have a tool that gives them an on-ramp to unlocking that potential within themselves, I think that speaks to every educator’s desire about, you know, what type of educator they want to be. I love that. I love that. And just this idea that we all want to get better. Teachers want to get better all the

time and do better by young people. And so, what you’re doing at Athena is helping with that. We like to close with this one-to-one concept. So, we’ll start with one person that is shaped your work or thinking that you’d like to share. Yeah, I think someone who’s definitely had an impact on my thinking over the years is Deborah Ball. She is a key researcher in this field of using video as a tool for teacher learning. She’s been a great advisor over the years. She’s now

the director of an organization called Teaching Works, which tries to understand what, how can you talk about the concrete practices that teachers can be working on to make that kind of immediate gain in their professional proficiency. But, you know, her research and guidance and advice of the years has just been really important to us and in the way we think about building tools that deliver on the promise of, as she says, decades of research.

Yeah, I appreciate that. And thanks for sharing about Deborah. So, two insights, well, other than that video does not have an opinion, which is something that I’m going to have to say somewhere at some point because I really like that. Two insights I’ve gleaned from this conversation. So, one, that this is not a tech challenge. The tech challenge has been solved. And of course, you’re introducing AI, this AI tool, which will continue to evolve. But it’s about organizational change

management, which we see in education over and over again, this mindset work and how can we help schools and leaders and teachers see that there are other ways to do things and video observation platforms are a way to do that. And then the second takeaway for me is that I love how you said coaching teachers to coach themselves. This ownership, this autonomy, this agency, it’s all about what we’re trying to do with young people, right? We want learners to be lifelong

learners and to take ownership over their own learning. And this is no different what you’re doing here. You’re trying to get teachers to take control of their own learning, which in this case is a coaching model. So, I really appreciate those two big themes kind of resonate for me in the bigger educational landscape. Adam, as we close out here, what would you add to that? What’s the takeaway message for you or a message you’d like to share with our audience?

No, I think the last thing I would add here is, you know, really just thinking about the reality that using video within professional learning is no longer an if, right? This is a question of when, when we use video as part of teacher learning. And so, if you are an educator who’s listening to this, who is interested in bringing to bear the best possible resources from a research perspective, right? This is not Adam jumping up and down because he’s passionate about Athena. This is what the

research says helps teachers improve and helps them have greater impact with students. When you use video, you get better outcomes in professional learning. You know, so that’s, that’s the other thing I would add here. Adam Geller, CEO of Athena. Thank you so much for your time today. I know that our listeners are going to be super intrigued and I hope you get some folks talking to you about the AI work you’re doing. And for all of our listeners out there, just keep thinking about that.

How do we help teachers coach themselves? And there’s, I know there’s so many teachers out here that are doing that well. Share abundantly with your colleagues. Pick up a camera or video and start watching each other teach. It’s an amazing process. Adam, just to finish off here, we’ll put a lot of notes in our, on our website when we release the pod. Where can they find out more about Athena? Best place. Best place www.edatheana.com. And if you want a shortcut to find information

about the AI coach platform, head to youraicoach.com. Adam, thank you so much for joining us, Getting Smart. Today, it’s been a pleasure and we’re excited that you’re able to share with our audience in the greater educational ecosystem. So good luck to you and we look forward to seeing your progress over time. Thanks for tuning into the Getting Smart podcast today. We want this podcast to be actionable and insightful and a great way to learn about what’s

next in learning. In order to stay on the cutting edge, we need people in the field to tell us what they’re hearing, what they’re wanting, and what they’re needing to learn more about. Got a topic or a guest in mind? Send your recommendations to me, Mason at GettingSmart.com. And if you like what you’re hearing, don’t forget to leave a review in Apple podcasts or subscribe wherever you listen. Feel free to share the podcast on social media using the hashtag GS Podcasts. Thanks so much.