Our advanced technology

At Sown To Grow, we believe in using technology only where it can support an educator's practice, and take pride in partnering with both educators and leading education research institutions to develop new technology.

Technology features

Concerning reflection alerts

Our technology uses natural language processing and machine learning to identify students who may need additional support.

The system reviews and flags reflections that may indicate a concerning emotional state. Each flagged reflection is manually reviewed by trained professionals before educators are alerted. This helps educators quickly prioritize students who may need an intervention outside of Sown To Grow.
An illustration of the Concerning Reflection Alerts System showing a list of student reflections that the system flagged as having concerning language or tone.

Responsive feedback for emotional reflections

Feedback suggestions act as a scaffold for educators who want support in responding to students’ social, emotional, and academic needs.

Sown To Grow’s advanced technology provides teachers with multiple suggested responses to a student based on the theme of their reflection. The suggestions help deepen student-teacher relationships and student reflection quality.
This work was developed in partnership with the National Science Foundation.
This work was developed in partnership with the U.S. Department of Education.
An illustration of the teacher response to student Weekly Emotional Check-In. Student expresses stress and negative feelings in their reflection. The teacher selects from response suggestions to help provide support to the student.

Learning strategies for academic reflections

Learning strategy recommendations help teachers suggest new ways for students to find what works for them.

Our technology leverages the latest learning science to identify the regulation, resource management, and cognitive strategies that students are using. From there, we recommend relevant new strategies for teachers to share with students.
This work was developed in partnership with the National Science Foundation.
A graphic showing how the response suggestions technology works. It shows how the technology reads a reflection, then categorizes the reflection, and outputs a suggestion based on that analysis.

Strategic reflection quality ratings

Strategic reflection quality score helps teachers prioritize and measure growth in student academic reflections, based on a research-informed framework.

Our technology rates student academic reflections on a rubric, developed in collaboration with educators. Ratings help teachers identify students falling behind and coach students to reflect more strategically on their academic goals.
An illustration of the Weekly Academic Check-In Teacher Portal view where a teacher can see a list of students in their class with each students' selected emoji for the week, the quality score for their reflection, the student's reflection, and a button for the teacher to respond to the student. Below that, there is a scale from 0-4 showing a range of less strategic to more strategic.

Using artificial intelligence (AI) with integrity

A green circle with a white chalk illustration inside it of a person at a blackboard at the front of a classroom with three people listening in the audience.
Educators always have oversight

When using AI to support teachers’ practice, we ensure an educator approves every suggested response before a student sees it.

A green circle with a white chalk illustration inside it of a hand facing up with a heart floating above it.
Act with transparency & trust

We are open with and responsive to educators about how our technology scores, categorizes, and makes suggestions on student reflections.

A green circle with a white chalk illustration inside it three stacked cylinders. The top cylinder has accent marks sparking out of it.
Be vigilant of unintended bias in data

Models are trained on data. If biases exist in the data, there will be biases in the model. We stay vigilant to potential sources of data bias.

A green circle with a white chalk illustration inside it of three smaller circles. One circle is filled in solid, one circle has polka dots in it, and one circle has stripes in it.
Leverage our team diversity

Our team brings a diverse set of backgrounds. We continually try to identify and fill our perspective gaps for model decision-making.

Have any questions?