AI as a Reading Coach: The Amira Learning Journey with Mark Angel

Mark Angel is the CEO and Co-founder of Amira Learning, the company bringing the science of reading and AI together. Amira is the first intelligent reading assistan for teachers and is now working with two million students in all 50 states and 14 countries.

💡 5 Things You’ll Learn in This Episode:

1. How Amira Learning uses AI to revolutionize reading education.
2. The importance of efficacy studies in edtech product adoption.
3. Strategic partnerships for scaling AI-driven tools in education.
4. Tackling privacy and equity concerns with emerging AI tools.
5. A vision of AI’s role in solving the global literacy crisis.

✨ Episode Highlights:

[00:02:15] The origin story of Amira, blending AI innovation with edtech experience.
[00:03:46] How the reading coach adapts to diverse learners, including ELLs and dyslexic students.
[00:07:40] The role of efficacy studies in building trust and demonstrating impact.
[00:09:53] Advice for districts navigating the surge of AI tools in classrooms.
[00:14:53] Optimistic vision of AI addressing the literacy crisis.
[00:18:43] Importance of scale and data in driving AI-based educational solutions.
[00:20:49] Key advice for entrepreneurs: focus on metrics and team culture.

😎 Stay updated with Edtech Insiders! 

• Follow our Podcast on:
  • Apple Podcasts
  • Spotify
  • Amazon Music
• Sign up for the Edtech Insiders newsletter.
• Follow Edtech Insiders on LinkedIn!

🎉 Presenting Sponsor:

Innovation in preK to gray learning is powered by exceptional people. For over 15 years, EdTech companies of all sizes and stages have trusted HireEducation to find the talent that drives impact. When specific skills and experiences are mission-critical, HireEducation is a partner that delivers. Offering permanent, fractional, and executive recruitment, HireEducation knows the go-to-market talent you need. Learn more at HireEdu.com.

Cooley LLP is the go-to law firm for education and edtech innovators, offering industry-informed counsel across the 'pre-K to gray' spectrum. With a multidisciplinary approach and a powerful edtech ecosystem, Cooley helps shape the future of education.

Transcript

[00:00:00] Mark Angel: I'll just say very, very openly and very, very straightforwardly, I don't think any of us have a right. to plunk technology into a classroom unless we have put it through the paces of rigorous random control trials, rigorous independent research. I was privileged to be able to walk in the footsteps of these great researchers from Carnegie Mellon, but as we've changed and evolved Amira, We have a responsibility to keep the research up to date and to keep proving that it works.

[00:00:38] Alex Sarlin: Welcome to EdTech Insiders, the top podcast covering the education technology industry. From funding rounds, to impact, to AI developments across early childhood, K 12, higher ed, and work. You'll find it all here at EdTech Insiders. 

[00:00:54] Ben Kornell: Remember to subscribe to the pod, check out our newsletter, and also our event calendar.

And to go deeper, check out edtechinsiders. org. Plus where you can get premium content, access to our WhatsApp channel, early access to events and back channel insights from Alex and Ben. Hope you enjoyed today's pod.

Today's conversation is with Mark Angel, the CEO and co founder of Mira Learning. It's the company that's bringing the science of reading and AI together. Amira is the first intelligent reading assistant for teachers and is now working with 2 million students in all 50 States and 14 countries. Enjoy our conversation today with Mark Angel, CEO and co founder of Amira Learning.

Hi everybody. We're back with the podcast and we were just talking three years ago when EdTech Insiders started. Man, we were excited about remote learning, but we had no idea that the AI revolution was about to hit the world. Well, somebody who did and who really has been leading the way on the AI and revolution AI in education revolution in ed tech is here with us, Mark Angel, CEO of Amira learning.

Thanks so much for joining us, Mark. 

[00:02:12] Mark Angel: Oh, my pleasure. Been looking forward to it. 

[00:02:15] Ben Kornell: So maybe Mark, just tell us a little bit first about Amira and the history and the origin story. 

[00:02:20] Mark Angel: Yeah, Amir is really a combination of two threads of learning. The first happened at Carnegie Mellon University, which has always been the hotbed for innovation and especially innovation around AI.

Unbelievably, they had the vision. That someday the tech would be smart enough to have a reading assistant for teachers come along, and they had that idea in 1993. So they spent Wow. Literally decades pursuing the notion that at some point we could use tech to help kids learn to read. The second thread was kind of my personal thread.

I've always been involved in. What people now call AI, going back for a long time, I helped to build some of the original speech recognition applications. And I wanted to get into education and ended up at a great company, Renaissance Learning, and what I experienced was the fact that the tech was not really ready to help kids do things like Learn numeracy and learn literacy, and so my failure at Renaissance and Carnegie Mellon's success in their Neck of the Woods came together to produce Samira, and so five years ago we started to work with the schools, today we're working with close to four million kids.

[00:03:46] Alex Sarlin: Incredible. And you mentioned a reading coach. We're in this sort of science of reading revolution in the U. S. where everybody's hewing to evidence based practices, and you do a number of different things. Your reading coach can be a dyslexia screener, it can diagnose, it can sort of work alongside a kid, it can support multilingual learners.

Give us a little bit of an overview of how the AMIRA reading coach works for different populations. 

[00:04:12] Mark Angel: It's great that you framed it that way because the central thesis of Amira is, is that the secret to helping kids become motivated, masterful readers is to be able to listen to them and to understand what their brain is going through as they try to learn to read.

And the key is, If we can understand what's happening in those neural connections as a student's learning to read, and we can understand the science behind their struggle, then yes, we can do all these amazing things like detect when they're dyslexic and understand when they're really struggling to comprehend.

So the way to think about Amira is, is that Amira is an AI powered listening. Engine for oral reading and then uses the superpower of listening to take other forms of AI to remediate the challenges that a student's reading brain is encountering. 

[00:05:15] Ben Kornell: You know, you made a early decision to partner with content providers and publishers and also to lean in on assessment as like a key leverage for success.

These are often, you know, for people who are outside of education, these are very subtle moves that actually can be really meaningful in the growth of the business and the impact. How did you decide to make those decisions? And, you know, you could have gone anywhere with Amira. Why did you make those bets?

And, and what advice would you have for others thinking about those same things? 

[00:05:50] Mark Angel: Well, Amir is my third, maybe my fourth time as a CEO founder. And I think one of the things that you learn when you're doing startups is you need to stick to what you're really good at and have the humility to understand that other people Are likely to be better at other things than you are.

And we wanted to be good at the AI. We wanted to be good at building the coach. A lot of folks are good at creating curriculum and content. So we just honored the fact that there's some great people in the education sector who understand instructional design, who understand pedagogy, who understand how to build engaging resources.

So we designed Amira to. Be able to use their work. And we stuck to what we knew other people couldn't do, which is to build an AI powered coach. And some of that kind of humility just generally drove our decision making. We knew it's hard to do direct selling to districts. So we've worked with partners in that dimension.

We knew that it's hard to help teachers engage and use technology with fidelity. So we've complimented the cool tech. Uh, with a relentless focus on success management and helping teachers to feel comfortable with, uh, how Amira works. 

[00:07:13] Alex Sarlin: The research that spawned Amira came, as you mentioned, out of Carnegie Mellon and a lot of really thoughtful research.

You've also pursued research on your own to look at the efficacy of the tool and seen some pretty, you know, outstanding results. Can you talk through. First off, how you decided to do this kind of efficacy study, sort of how it played into your overall strategy, and then what actually came out of it and what it's meant for you since you've gotten the results.

[00:07:40] Mark Angel: Well, thanks for that question, because I'm a radical in one way. And that way is believing that we don't have a right to put five and six year olds on to unproven technology and unproven interventions. One of my previous efforts was in the medical space, and I learned there that You know, you don't experiment with people.

You, if you've got a new idea for a new drug or a new treatment, you spend years putting it through clinical trials, getting approvals, demonstrating that not only does it work, but it doesn't have negative side effects. Somehow all that rigor and all that respect for the user has not really penetrated into the education space.

Remarkably, because we. All are in this space because we care about the kids. So I'll just say very, very openly and very, very straightforwardly. I don't think any of us have a right. To plunk technology into a classroom, unless we have put it through the paces of rigorous random control trials, rigorous independent research.

I was privileged to be able to walk in the footsteps of these great researchers from Carnegie Mellon, but as we've changed and evolved Amira, we have a responsibility to keep the research up to date and to keep proving that it works. Great news. It is. Absolutely amazing. The effect sizes that Amira is producing today.

And it's the sole and legitimate reason why we've rocketed to the millions of kids on the software that we have. 

[00:09:26] Ben Kornell: Yeah. I'm very curious implied in that statement is also like an understanding that many AI enabled companies have not met that bar and we also have the entrant of generalized. AI, which is not tuned, especially for under thirteens.

How do you think schools or districts should navigate this kind of Cambrian explosion of AI tools and apps? And how do you think that field will evolve? 

[00:09:53] Mark Angel: Well, I've talked to a lot of superintendents who rightfully say that they do not want to repeat the experience with social networks, right? We saw an amazing, brilliant.

Innovative technology come into the world. That technology has changed much of how we live and work. It's changed politics. It's changed the way news is delivered. It's changed how we interface with our families, but sadly, it also changed the way that we live and work. That kids communicate and they learn and often not for the good.

And so I think informed educators are thinking about that experience and saying, never again. And so hopefully, uh, we will have learned our lesson and we will understand. That the things we put in front of children need to be fit for purpose. They need to be tested. They need to be thoughtfully deployed and they need to be under the control of the people who are responsible for the children, meaning the school districts.

Right. We shouldn't be in a spot where kids are migrating onto public sites and having Data exchanges and interfaces that are completely out of the control of the school districts who are supposed to be monitoring, supervising those children during school hours. So I'm hopeful, optimistic that we'll learn from experience and that we'll do a better job this time of taking a powerful, powerful tool and using it the way that does some good instead of does some harm.

[00:11:31] Alex Sarlin: I love your phrasing of Amira being a listening, you know, engine that it's really about that the listening is its superpower. And that's, I think it's really interesting. And it's, you know, as we sort of enter these new types of interfaces for AI, especially voice, you've been in this space for a while.

I'm curious when you first started talking to schools, was there any pushback on the idea of, you know, listening to students voices or analyzing students voices? Pronunciation or their reading skill or their reading speed as a sort of privacy issue. And if so, has that gone down over time, especially with the effect size.

I'm also asking partially because I think some of the entrepreneurs who listen to this are thinking about, you know, listening and what it means in schools. 

[00:12:15] Mark Angel: Yeah, it's a great question. We've always had pushback. We still do today. We rightfully have pushback on the privacy front. We rightfully have pushback on the equity front.

And those are both important. And I think any responsible vendor has to step up to the, Risks and difficulties that are associated with the innovation we're trying to create and, and, and, and utilize. To your point, the fact that millions of kids and close to 2, 000 districts in all 50 States are using Amira definitely takes some of the worry and concern out and five years after we've been in schools, the fact that there's never been, Uh, a single incident related to privacy or a single problem that has caused people to regret using a mirror has helped us tremendously.

But on the flip side, as people become more aware of AI and they understand some of the general risks, it's also heightened the kinds of questions that we get. And this again goes back to the fit for purpose situation. We're able to show school districts that we handle data in exactly the same way that their trusted, traditional ed tech partners handle data.

And at the end of the day, there's simply no difference if we're doing our job and they're doing theirs, we can protect the privacy of the children. And we can demonstrate that we can deliver a mirror's capabilities, not only The research has shown that Amira's largest effect size are in the kids that I personally care about most.

The kids in our big cities. The kids who are English language learners. The kids who are dyslexic and have struggles. They're the ones who have shown the biggest gains. So, Amira's actually a little inequitable, but in exactly the right way. So, 

[00:14:12] Ben Kornell: So, Mark, you know, your career as an entrepreneur has spanned many eras.

Of just moving from paper to digital to now digital to AI and adaptive. And as you mentioned at Renaissance, you know, one of the realizations was the, the technology was not necessarily ready to meet the moment for what educators and learners need. Do you believe that AI now is that technology for that moment writ large?

And as you fast forward a decade from now. What do you imagine schools and learning environments could look like given the window of insight you've had at Amira? 

[00:14:53] Mark Angel: Yeah, well, I want to start by saying that if you had asked me five years ago, When we started Amira, if we would see the kinds of generative AI capabilities we see today, I have never imagined that things would have come so far so fast.

So I want to say in advance that my crystal ball is always pretty murky when it comes to the pace of technology. Sometimes it comes at us faster than we can ever imagine. Sometimes with self driving cars and Things like that, it comes at a slower than we can probably ever have imagined. So not sure about the exact pace, but I am 100 percent convinced and have been for a long time that AI is the game changing answer to the learning crisis that we see in this country and across the world, right?

When you look at what's happened in the medical space, again. By 1970, we had elongated people's lives tremendously, and yet since 1970, we've added 10 to 12 years to the life expectancy of every American. If you look at reading scores since 1970, they are exactly the same. Not one iota of difference. I promise you that 20 years from now, we will look back and we will see the fork in the road that came with AI, and we will see an acceleration in reading scores and outcomes that we look for and never got over decades.

So it's the right thing and the right time to solve this incredibly important. 

[00:16:40] Alex Sarlin: It's a very inspiring vision. And I completely agree. I think we, you know, reading. Our inability to successfully teach reading at scale in the U. S. is one of our, I think, national shames, really. And it is so exciting to see technology coming into play that may continue to, I mean, you have incredible people, don't get me wrong, incredible educators, incredible reading specialists, literacy experts all over the country working so hard to get this to work, but it is such a hard problem to do it in a country of our size, so I would love to see that.

Fork in the road moment that you're talking about. 

[00:17:14] Mark Angel: Yeah. And listen, it's important to say that as much as tech has, has had these incredible impacts on health and medicine, that impact has come because of a successful partnership with people, with doctors, doctors have not gone away. Teachers are not going to go away.

All AI is going to do is knowledge and power. And. Accelerate and unleash the magic of teachers the way the tools like magnetic resonance imaging machines and cancer fighting drugs have unleashed the power of surgeons and doctors, right? It's just, it's, it's, it's about the synergy and sapatico capabilities of the tech and the people that's going to get us there.

It's not one or the other. 

[00:18:05] Alex Sarlin: Completely, yes. A false dichotomy that a technology or a teacher, totally agree. Right. You mentioned something that I think would be very interesting for some of the listeners of the podcast. So Amira, as you mentioned, fit for purpose, very specific use of AI, very carefully trained AI, lots of research behind it.

And then you've worked with some really big and interesting partners to help scale the technology, including Hod Mifflin Harcourt and iStation, which you merged with this year. You tell us a little bit about that journey as a serial entrepreneur. What do you look for in that kind of partner and why is it really helpful in growth and success for Amira?

[00:18:43] Mark Angel: Great question, Alex. I don't want to be overly simplistic, but I think sometimes what you learn as an entrepreneur is, is that complexity is your enemy. And I have tried to follow a really simple mantra around a mirror strategy, which is that scale is everything. And the key to AI is that. Data. Why is open AI been able to forge ahead of the Googles and the metas of the world?

They've been incredible at aggregating and marshaling data for the purposes of building their foundation models. So we've made a bet at Amira that. We're going to create student growth by marshaling as much data about how students learn as possible, and we follow the strategy that relentlessly allows us to seek scale, uh, as rapidly as we can.

And so Hodman from Harcourt was a huge accelerator for us. We expect that our combination with iStation will be a huge accelerator and we'll continue to push that. We're just announcing a partnership with Pearson in Canada. We'll pursue strategies that allow us to scale on a global basis so that Amira continues to have access to billions and billions of interactions with students.

[00:20:01] Ben Kornell: This has been an incredible conversation, not only because it's inspiring about what you've built and the future, but also the practical evidence that you have, that it is really moving the needle for learners. And I think, you know, as much as we get caught up in these hype cycles. Back to what you said, this is what we care about is unlocking the potential of every learner.

And so I guess for those entering the education conversation and AI conversation as the newest entrepreneurs, I'd love just some closing thoughts or advice that you would have as you're launching your company. What, you know, given your experience, what would you advise them and how should they navigate this period of great potential, but also great change?

[00:20:49] Mark Angel: Yeah. I think whether it's education or any other space, successful companies discover. The one metric that really matters to them, and then they get incredibly focused on being good at moving that metric and the companies and entrepreneurs I admire all fall into that pattern of thinking not about the product, not about the financials.

But thinking about the change that you're trying to induce with your company and your mission. So that's number one, have a metric that really exemplifies the mission and the change you're trying to achieve with your, with your startup and your initiative. Second thing, and this is so trite, but it's so true.

Amira has gotten to this amazing place. Because of the commitment and talent level, the people who committed to the company. And so we always say, especially in software, it's all about the people, but it really is all about the people. I show up for work every day with amazing data scientists, with amazing success managers, with amazing co founders, and they.

Do the hard work every single day of trying to make Amira better and trying to make sure that the districts get the optimum value out of the product that we produce. So the more time you can spend on your team and the more time you can spend on culture, the more you're going to have a great chance of making that metric move.

[00:22:25] Ben Kornell: Awesome. Well, thanks so much for joining us. Mark Angel, CEO of Amira Learning. It's been wonderful. Thanks for joining us. 

[00:22:32] Mark Angel: Oh, my pleasure. This has been great. 

[00:22:34] Alex Sarlin: Thanks for listening to this episode of EdTech Insiders. If you like the podcast, remember to rate it and share it with others in the EdTech community.

For those who want even more EdTech Insider, subscribe to the free EdTech Insiders newsletter on Substack.