Introducing: Sana Labs

Just as the best tech entertainment companies – Spotify, Netflix, Apple Music – are attempting to provide users with richer personalized experiences by tailoring content and recommendations to the unique profile of each individual user, so are Sana Labs aiming to apply similar principles to the world of education. Their AI-powered solution offers real-time content recommendations for forward-thinking education companies around the world, and we caught up with their CMO and co-founder, Anna Nordell, to learn more…

Anna Nordell

CMO and co-founder

Sana Labs

Sana Labs uses breakthroughs in AI/ML to model content, model students, and optimize strategies to maximize each student’s learning outcomes by increasing engagement and proficiency in every single interaction. The company has developed a solution that is easily integrated into digital educators (publishers and established edtech companies that offer interactive digital learning platforms), making their courses fully personalized and adaptive to each student’s learning patterns, knowledge level, knowledge gaps and more.

Currently, the Sana platform offers two leading features designed to empower the student, teacher and classroom:

  1. Sana Learn – Empowering Students

Sana Practice recommends continuously personalized learning materials for students. The live recommendations identify, using machine learning techniques, exactly what each student should study in every interaction to optimize his or her learning outcomes.

  1. Sana Insights – Empowering Teachers

Sana Insights provide educators and teachers with invaluable insights into student learning. This information assists the provision of differentiated instruction and timely interventions that are truly hands-on and tailored to teachers’ needs (developed together with teachers).

Sana’s technology makes innovative use of recurrent neural networks and reinforcement learning. These models take content, answers, response times and an array of contextual information as inputs, before processing these inputs through a number of different network layers containing millions of neuron-like connections. One neural network, – the “prediction network” -, predicts how the student will interact with content. Another -, the “planning network” – , selects the content that will result in the highest increase in engagement and proficiency over time.

The model incorporates features about the student such as proficiencies, assessment needs, need for remediation, and where to move forward;, about the response and behavior such as pace, repetition, engagement and preferred strategies as well as about the content such as asset relationships, instructional value, assessment value, difficulty, and duration. Capturing a rich and accurate representation of students and content from hundreds of millions of data points.

Personalized education is proven to produce learning gains for the average student to the order of two standard deviations. Our mission is to apply recent breakthroughs in deep learning to make personalized education accessible to everyone — providing these benefits on a global scale.

Currently, Sana Labs is working with digital education companies around the world with over 70 million students in their platforms. Consequently, Sana Labs has access to massive data-sets on how people learn, creating immense data network effects as well as a thorough understanding of how to best train models and algorithms to optimize each student’s learning outcomes.

  • World-class engineers and researchers – At Sana Labs, we are a team of world-class engineers and researchers with an extensive experience of putting algorithms in production, within a vast amount of fields. Our teams’, backgrounds include experience at Imperial College, CERN, Google, and Spotify.
  • Unbeatable Time to Market – Our API makes integration with any architecture and content simple so that personalization features can be production-ready within days, not months.
  • Best in Class Results – Sana Labs has received independent recognition for its solution, including 1st Place in the Duolingo Global AI Benchmark Challenge, and 1stplace in the ASSISTment Global AI Benchmark Challenge.

The global education industry is vast (valued at over $6 trillion in 2016), and is undergoing a sea change through the move to digital and online instruction, materials and modalities. This shift is enabling a progression towards what many educators consider to be the holy grail of learning — universal personalized, adaptive instruction and assessment. With our solution, we aim to be the engine that drives forward this change — fundamentally improving the function of the entire industry, directly affecting hundreds of millions of people’s lives every day through personalizing education at scale.

Anna Nordell is the CMO and co-founder of Sana Labs. To meet their CEO and co-founder, Joel Hellermark, readers are warmly invited to join the 2018 Reimagine Education Conference, at which, alongside IBM representatives, Joel will be speaking on the 30th November.