K-12 Playground – AWS Innovation Islands

STEM Rover AI Coding Companion

Coding education in middle and high school faces a scaling problem: meaningful learning requires one-on-one guidance, but teachers cannot provide individualized mentoring to every student simultaneously. General-purpose AI tools sound promising but come with many challenges as they readily hand students complete answers rather than teaching them to think through problems, lack awareness of specific hardware and project constraints, and offer no guardrails against content that violates school acceptable use policies.

Deloitte’s Believers Program partnered with the AWS Cloud Innovation Center (CIC) at Cal Poly to build an AI coding tutor powered by Kiro, AWS’s AI-powered development environment. The Believers AI Tutor lives directly inside the student’s programming environment. It provides real-time, conversational guidance as students program physical robot rover kits using Python on Raspberry Pi hardware. The tutor acts as a Socratic guide: when a student asks how to make their rover turn left, it explains GPIO pin control and motor direction, then asks the student to try writing the code themselves rather than handing them the solution.

What makes this platform uniquely suited for education is how teachers govern the AI’s behavior through Kiro steering files — plain-language files that shape the tutor’s responses without requiring any coding or cloud configuration. Teachers can:

Define the AI’s role as a tutor that guides rather than gives answers.
Enforce code snippet limits and blocks solution extraction
Implement the school district’s acceptable use policies
Provide the rover’s complete hardware specification so the AI gives technically accurate, device-specific advice.

In this demo, see how students and teachers interact with the Believers AI coding companion.

Diverse Learner Support

The K-12 Co-Teacher is an AI-powered assistant that streamlines the management of student accommodations and support plans for educators. By integrating IEPs, 504 plans, and psychological reports with daily lesson planning, the solution reduces administrative burden while ensuring consistent support for students. The system provides real-time accommodation surfacing, class-level and individual student views, and maintains living records of student progress. Through intelligent analysis of lesson content, it automatically flags potential challenges and recommends inclusive strategies aligned with documented supports, making differentiated instruction more manageable for teachers dealing with multiple class sections.

AWS services utilized:

Amazon Bedrock (Anthropic Claude Sonnet 3.7)
AWS Amplify
Amazon Cognito
Amazon API Gateway
AWS Lambda
Amazon DynamoDB
Amazon API Gateway WebSocket APIs

This solution transforms classroom support management by reducing teacher workload, improving consistency of accommodations, and enabling more time for direct student engagement.

Document Processing with Bedrock

Despite operating in a digital-first world, most K-12 institutions still struggle with manual processes that rely on paper forms and documents. Leveraging AI-powered Intelligent Document Processing (IDP), schools can significantly reduce the time and effort required for manual data entry and processing, streamlining tasks such as student registration, student cumulative folders, health records management, and more.

Services like Amazon Bedrock and Amazon Textract can accurately extract text, handwriting, and information from scanned documents and forms. Once extracted, this data can be processed in a variety of ways, including classification, analysis, summarization, redaction, human review, or integration with other systems and applications.

In this demo, see how generative AI and services like Amazon Bedrock can be used to quickly create Intelligent Document Processing pipelines for K12 documents.

Enrollment Prediction

Higley Unified School District serves approximately 13,500 students across 16 schools. Predicting school enrollment is crucial.

According to local school administrators, simply getting static predictions can cost as much as $20,000, an amount that can be a significant burden for the school districts and does not have the flexibility to accommodate changing circumstances.

The ASU Cloud Innovation Center created an open-source prototype to predict future enrollment and integrate real time data, such as attendance data to assess enrollment predictions. This solution uses a wide pool of data from various sources and provides analytics with an easy to use dashboard.

Open-source code: Download

Generative Business Intelligence

For data analysis and visualization, many schools today still rely on pre-defined data models and dashboards that can be time-consuming to create and update and difficult for users to navigate and use. Using generative AI, both dashboard authors and consumers can streamline access to insights and visualizations with self-service, natural language queries.

Leveraging Amazon Q in QuickSight, users can pose questions using natural language, and the service will interpret the query, retrieve the relevant data, perform necessary calculations, and generate visualizations to provide insights. This approach eliminates the need for users to navigate through predefined dashboards or write complex SQL queries, making data exploration more accessible and efficient. Additionally, Amazon QuickSight can help users quickly generate executive summaries, narrative data stories, forecasts, and what-if scenarios.

In this demo, see how users can leverage Amazon Q in QuickSight to ask their own natural language questions of K-12 education data.

Document Processing with Textract

In this demo, see how Intelligent Document Processing can be used with documents required for school enrollment.

Student Assistant

Generative AI can provide personalized assistance and tailored feedback to students, enhancing the student learning experience and fostering independent problem-solving skills. Leveraging the power of artificial intelligence, generative AI can adapt to individual student needs and interests, providing customized explanations, examples, and exercises based on learning style and level of understanding.

With services like Amazon Bedrock, developers can easily create and deploy generative AI applications tailored to specific subjects and topics. These models can be augmented with existing data like educational content, textbooks, lecture notes, and student information, enabling them to provide more accurate and relevant assistance to students. With Amazon Bedrock, you have full control over the security and privacy of your data, and can implement guardrails to ensure safe and responsible AI use.

In this demo, see how some of the core features of Amazon Bedrock, like Agents, Knowledge Bases, and Guardrails, can be used for a student assistant.

District Chatbot

Schools are often flooded with requests for information from parents, employees, and community members. Generative AI chatbots can enhance the user experience and reduce burden on school staff by addressing frequently asked questions and automating common requests or workflows. These chatbots can provide a seamless and intuitive interface in multiple languages that lets district staff focus their efforts on more complex tasks that require human intervention.

Services and solutions like Amazon Bedrock, Amazon Lex, and QnABot on AWS can be used to implement conversational chatbots that are capable of interpreting and understanding user intent to provide relevant and accurate responses based on your unique organizational data and information. Combined with Amazon Connect, these chatbots can be embedded in various communication channels enabling users to interact through both voice- and text-based interfaces.

In this demo, see a sample FAQ chatbot embedded on a district website.

IT & Developer Productivity

K-12 technology teams have to manage overwhelming workloads with countless day-to-day tasks. Generative AI can play a crucial role in improving IT productivity and efficiency, assisting staff and developers write and test code, operate applications, review security issues, and optimize solutions.

Amazon Q Developer is an AI-powered coding companion that can integrate with popular IDEs and understand your unique codebase to help developers write, analyze, and improve code inside or outside of AWS. Amazon Q Developer can also help you understand, build, extend, and operate applications on AWS; you can ask questions about your own AWS resources, best practices, documentation, and more.

In this demo, see how Amazon Q Developer can be used to generate unit tests, review code, and remediate issues.

Employee Assistant

In today’s fast-paced education environment, teachers and staff can struggle to find the information they need quickly, hindering their ability to focus on providing quality education for students. Generative AI can help by providing faster access to organizational information, policies, and knowledge.

Utilizing services like Amazon Q and Amazon Bedrock, staff can engage in natural language interactions to ask questions and receive relevant information from the organization’s own knowledge bases. Amazon Q and Bedrock can be trained or augmented with organizational documents, policies, and procedures, allowing it to understand unique context and provide accurate and tailored responses. This ensures employees have access to the most up-to-date and compliant information and saves time that is better spent with students and colleagues.

In this demo, see how Amazon Q for Business can answer questions based on unique organizational knowledge like a parent student handbook, school calendar, and acceptable use policy. Also see how Amazon Q Apps let users create, use, and share their own generative AI-powered apps for common, repeatable tasks.