Podsie

Despite decades of learning science breakthroughs on how students best learn and retain content, these best practices are widely underutilized in K-12 classrooms around the world. As a result, teachers often use suboptimal teaching strategies, and students, especially those who are traditionally underserved, are not being given their best chance to reach their full academic potential.

We're bridging the gap between the science of learning and K-12 classrooms. To start, we've built a web app that captures these best practices and makes them easy for teachers and students to use. For example, "spacing" is a learning strategy that has been proven to improve learning by 15%. It's historically been underutilized because it's difficult for teachers to implement manually. Podsie automates "spacing" for students in a personalized way.

Scaling Podsie out globally would drastically improve student learning outcomes and teacher pedagogy.

The majority of primary and secondary school classrooms around the world are not implementing best practices that are prescribed by learning science researchers. In order to help students recover from their learning loss due to the Covid-19 crisis, we should listen to scientists and widely adopt their research-backed learning strategies. 

These are the current barriers that make it difficult for classrooms to implement these learning strategies:

1. Lack of teacher training - A survey conducted by Deans for Impact found that out of 1,000 future teachers in the U.S., only 6% understood the concept of "retrieval practice", which is a strategy that has been proven to improve student learning by 150%. 

2. Difficult to implement manually - These learning strategies are typically difficult for teachers to implement manually, especially in traditionally underserved classrooms where students have a wide range of academic needs.

We built a web app that makes it easy for teachers and students to leverage three learning science best practices on a day-to-day basis: spacing, retrieval, and interleaving. 

On the Podsie web app, teachers assign questions for students to practice. Once students are done with the questions, each question is inserted into a student's personal deck, which is a feature that uses a spacing algorithm to personalize when each student should review a question again based on their prior performance.

For example, if Johnny misses question A, but gets question B correct, then Johnny will be prompted to practice question A the next day, but he won't be prompted to practice question B until a week has passed. This way, each student will only focus on reviewing content that they need to work on.

Our platform is built on Ruby on Rails on the backend and uses React.js on the frontend. 

The target population we're serving are students and teachers at Title 1 schools (40% of students qualify for free and reduced lunch) in the US. Teachers and students at Title 1 schools are traditionally under-resourced, and the Covid-19 crisis has disproportionately affected them. Our solution directly help these students recover from the learning loss that was caused by the pandemic by providing them with the most up to date learning best practices.

To better understand their needs, we've worked closely with teachers at Title 1 schools during our pilot study. We regularly met with these teachers to gather feedback about our web tool, and we also conducted surveys with their students to ensure that we were consistently building around their needs. In March, we partnered with a learning scientist at Carnegie Mellon to conduct an experiment that measured the efficacy of our web app in an eighth grade science classroom at a Title 1 school. We're still in the process of analyzing the data, and we plan to constantly run these studies in order to ensure that we're effectively and equitably serving our students and teachers.

In order to tackle the learning loss that was caused by the Covid-19 crisis, we need to rely on evidence-based pedagogical practices at scale. Podsie's web tool provides an easy way for teachers to learn about these practices and implement them in their classrooms. Our target population is students at Title 1 schools, where there is often a large range of academic needs among students. With this in mind, we've taken great care to build our tool so that it facilitates these evidence-based practices in a way that's personalized to each student's needs.

We launched our pilot in August of 2020, and over the past school year we've worked with 20 teachers and 800 students. The majority of teachers and students are at Title 1 schools in the US.

In this past year, we've spoken with several learning scientists, and their consensus was that the majority of existing edtech apps are not grounded in evidence-based practices. Podsie's approach is innovative because everything we do is built on the best practices that learning scientists have proven to be highly effective. For example, learning scientists have conducted studies to show that "spacing" can improve student retention by 15%, while "retrieval" can improve student retention by 150%. Capturing these learning gains at scale would drastically improve student learning.

In addition, because Podsie is grounded in research, we also run experiments to measure our own efficacy, and then we transparently report the outcomes. We think that concretely measuring and demonstrating our impact will influence other edtech companies to become more learning outcomes-focused, to conduct studies to measure their own efficacy, and to transparently report the impact of their products.

We're currently wrapping up our pilot study, where we served 20 teachers, and around 800 students. We're launching our app publicly in the next couple of weeks, and we plan to grow to serve 2,000 teachers and 100,000 students in one year. In our first year, we plan to focus on classrooms in the United States.

In five years, we plan to serve 100,000 teachers and 5 million students, and by then our goal is to expand to English-speaking countries around the world.

The indicator for the scale/reach of our web tool is based on daily active teachers and students. 

We run expriments to measure for how much we're improving student learning outcomes. For example, we're currently in the middle of an experiment where we're comparing how much content students are able to retain if they use Podsie vs if they do not use Podsie. Our goal for the next school year is to run a large-scale RCT to further measure our app's efficacy in improving student learning.

2 full-time staff, 1 part-time staff

Podsie's CEO, Josh, taught for two years under Teach For America, has a Master’s in Education from Stanford, and is currently a volunteer AP Comp Sci teacher for a local high school. In 2016, Josh transitioned into software engineering, so he can directly use his deep understanding of the education landscape and the needs of teachers and students into building a tech product that teachers and students love.

Podsie's CTO, Jesse brings a unique blend of both product expertise (former Facebook and Airbnb software engineer) and passion about educational impact.

Podsie is a nonprofit edtech company, and Jesse and Josh come from backgrounds that are well-represented in tech (Asian American male from middle class upbringing). However, DEI matters to us, and we’re actively trying to make Podsie more diverse, equitable, and inclusive. The first step we’ve taken was to ensure that the majority of our board of directors come from backgrounds that are traditionally underrepresented in tech. Concretely, this means that 2 out of our 3 board members are women. As we grow, we plan to recruit leaders who come from different backgrounds, especially people of color who are underrepresented in the tech industry and who come from backgrounds similar to the students who we focus on serving.

We mentioned that one of the key barriers we're tackling right now is fundraising. Beyond the access to funding in grants and investments, we highly value the mentorship, coaching, and strategic advice that Solve offers. As part of Solve's network, we think we'll be able to learn a lot about how to best position ourselves for long term sustainability.

More importantly, we're applying to Solve because the Equitable Classrooms challenge resonated deeply with us. In order to help students recover from the unfinished learning that was caused by the Covid-19 crisis, we must listen to scientists and rely on evidence-based strategies. More importantly, we need to ensure that the most underserved has easy access to evidence-based strategies in order to accelerate their educational recovery. Finally, we think our solution aligns well with Solve's call for a solution that "support[s] teachers to adapt their pedagogy" and "facilitate[s] personalized instruction".

Financial - As mentioned in the barriers question, we're currently fundraising, and we'd value help in pitching to philanthropists and donors.

Product / Service Distribution - At the moment, we've primarily focused on distributing our web tool directly to teachers. However, we'd eventually be interested in working with school districts, and would highly value advice from someone who has worked with school districts before.

We'd like to partner with the Bill and Melinda Gates Foundation. One of their key areas of focus is helping with teacher preparation, particularly for teachers who teach students experiencing poverty. Another one of their focus is to "accelerate the integration and translation of evidence-informed teaching practices and research into scalable education practices, programs, and tools that support teachers and students." Both of these focus areas align closely with Podsie's mission and goals, and we'd love to work with them to get connected with other organizations who are also tackling these same problems.

We'd also like to partner with Professor Justin Reich, director of the Teaching Systems Lab at MIT. Professor Reich's book, Failure to Disrupt, was hugely influential on our work and our philosophy of continually experimenting and measuring the efficacy of our solution. We'd highly value Professor Reich's advisorship.

Podsie is qualified for this prize because:

1. Our mission is to increase equitable access to quality education that's grounded in evidence-based strategies.

2. One of our key focus areas is teacher professional development and helping teachers learn more about practices that are grounded in the science of learning.

3. Most importantly, Podsie is also committed to helping students know themselves and understand how their brains learn effectively. We believe that this "understanding how the brain learns" is an invaluable asset that will have a lasting impact on the rest of a student's academic trajectory.

We'd use The ASA Prize for Equitable Education to fund our teacher professional development efforts. We plan to put together workshops and develop more teacher training resources.

Podsie leverages data science in order to make accurate and personalized recommendations on which areas of studies each student should focus on. Through this, we're able to provide a personalized learning experience for students that maximizes how they spend their study time. We also rely on Natural Language Processing to help evaluate and grade student responses.

Our team is currently trying to optimize the algorithms that we use to make student recommendations. In order to do this, we're looking to hire a data scientist, and we'd use The AI for Humanity Prize to fund this hire.

Podsie leverages data science in order to make accurate and personalized recommendations on which areas of studies each student should focus on. Through this, we're able to provide a personalized learning experience for students that maximizes how they spend their study time. We also rely on Natural Language Processing to help evaluate and grade student responses.

Our team is currently trying to optimize the algorithms that we use to personalize each student's learning experience. In order to do this, we're looking to hire a data scientist, and we'd use The AI for Humanity Prize to fund this hire.

Podsie is qualified for this prize because we're using web technology to expand access to highly effective, evidence-based learning strategies that have traditionally been difficult for teachers and students to leverage in classrooms. We also leverage data science in order to provide personalized learning experiences for students.

Our team is currently trying to optimize the algorithms that we use to personalize each student's learning experience. In order to do this, we're looking to hire a data scientist, and we'd use The GSR Prize to fund this hire.