AWS Education Podcast

S1E26: 26: Bridging Academia & Industry; Shaping AI-Ready Talent

In this episode, Pratik Chunawala shares his experience bridging the gap between academia and industry through deep partnerships with NYU Tandon, Carnegie Mellon University, and the City University of New York (CUNY). As a Principal Consultant in AWS ProServe, Pratik discusses how he's helping shape curriculum to prepare students for the AI era, moving beyond tool usage to governance, security architecture, and spec-driven development. He also shares his bold 2031 prediction for where AI is headed. Key Discussion Points * Serving on the NYU Tandon Alumni Advisory Council to mentor women in STEM and underrepresented students * Visiting lecturer role at Carnegie Mellon University connecting curriculum to industry best practices * New appointment to CUNY's AI and Emerging Tech Advisory Committee * Evolving curriculum from "learning to build AI" to "learning to govern AI in mission-critical systems" * Hands-on industry sessions bringing NYU students into Amazon's Bryant Park office with executive speakers * Spec-driven development with Kiro as a teaching methodology for real-world problem solving * Kiro for Students program: 1 year of free access with 1,000 monthly credits, no credit card required * The 2031 prediction: AI becomes invisible infrastructure, like electricity or the farming tractor Featured Technologies * Kiro (Agentic IDE) * Spec-Driven Development * Agentic AI Key Takeaways * Curriculum must shift focus from how tools work to how to apply tools to industry-specific problems, tools will be outdated by graduation * Spec-driven development teaches students to work backwards from problems, using steering documents rather than white coding or basic prompting * Kiro for Students lowers the barrier to entry with free, usage-based access across 11 initial universities including NYU * The most underrated professional skill: communicating technical decisions to non-technical stakeholders with competing incentives * Industry-academia partnerships give students exposure to real failure modes, interoperability challenges, and system security considerations Tags #GenAI #EdTech #AWS #AgenticAI #AmazonKiro #HigherEducation #AIinEducation #SpecDrivenDevelopment #WomenInSTEM #StudentSuccess #UniversityPartnerships #FutureOfWork #CurriculumDesign #NYUTandon #CUNY #CarnegieMellon #AWSProServe #KiroForStudents #IndustryAcademia #DigitalTransformation

S1E25: 25: Building Security Champions & Educating Teams for the AI Era

In this episode, Sarah Currey shares her insights on building a culture of security through AWS's Security Champions program and the innovative approaches to AI security in the age of autonomous agents. She discusses the launch of the free Security Champion Knowledge Path on AWS Skill Builder, AWS's partnership with Mido Academy in South Africa, and critical considerations for securing agentic AI systems in education and beyond. Key Discussion Points * AWS Guardian's Program with 5,000+ security champions driving velocity and security outcomes * Launch of the free Security Champion Knowledge Path on AWS Skill Builder with Credly certification * Partnership with Mido Academy in South Africa creating cybersecurity pathways out of poverty * Emotional intelligence and soft skills as critical differentiators in the AI automation era * Agentic AI security challenges and the AWS Agentic AI Scoping Matrix framework * Shift-left security approach and embedding security throughout application lifecycle Featured Technologies * AWS Skill Builder * AWS Security Reference Architecture * Agentic AI Scoping Matrix * Gen AI Scoping Matrix * Amazon Bedrock * MCP Server Integration Key Takeaways * Security champions increase product launch velocity by embedding security expertise across development teams * The Security Champion Knowledge Path offers free training with interactive Simulearn simulations and Credly certification * Mido Academy achieved 72% internship placement and 38% conversion to permanent employment in 2024 * Agentic AI systems require fundamentally different security approaches than traditional cloud models * Four critical questions for AI security: What can the agent access? What actions can it take? Does it require human approval? What's the scope of impact? Resources Mentioned * AWS Skill Builder Security Champion Knowledge Path * Credly Badge Certification * AWS Agentic AI Scoping Matrix * AWS Gen AI Scoping Matrix * AWS Security Reference Architecture for AI * Mido Academy Partnership * Women in Cybersecurity * The Cyber Guild Tags #AWS #CyberSecurity #SecurityChampions #AISecurity #AgenticAI #EdTech #HigherEducation #SkillBuilder #SecurityTraining #MidoAcademy #SouthAfrica #CyberSecurityEducation #CloudSecurity

S1E24: 24: Resiliency in Gen AI Applications

Episode Details  * Date:  April 13, 2026 * Duration: ~26 minutes  * Speakers: Pranusha Manchala (Host), Joe Chapman (Principal Solutions Architect at AWS)  Episode Summary  In this episode, Joe Chapman joins Pranusha Manchala to discuss the critical importance of resiliency in generative AI applications for education. Joe shares his expertise on building highly reliable and resilient AI systems, exploring how EdTech companies can ensure their AI-powered platforms remain available and trustworthy when students, teachers, and administrators need them most. The conversation covers shared fate architecture, fault isolation strategies, monitoring best practices, and actionable steps for implementing resilient Gen AI systems.  Key Discussion Points  * Evolution of EdTech from cloud migration to COVID-era scaling to Gen AI integration  * Why availability is non-negotiable for AI-powered learning platforms  * Understanding shared fate and blast radius in Gen AI architectures  * Fault isolation boundaries and hard vs. soft dependencies  * New monitoring dimensions specific to Gen AI systems  * Resiliency as a continuous journey, not a one-time implementation  * Practical testing strategies for Gen AI workloads at peak utilization  Featured Technologies  * Amazon Bedrock  * Multi-region inference  * Reasoning models  * AI agents and tools  * Knowledge base systems  Key Takeaways  * Students and teachers work on critical deadlines (11 PM before midnight submissions), making 24/7 availability essential  * AI implementations showing 10+ points higher accuracy require resilient infrastructure to maintain trust  * Five pillars of resilient systems: redundant components, sufficient capacity, timely output, correct output, and fault isolation  * Gen AI-specific metrics include reasoning traces, tool invocation patterns, and response quality baselines  * Amazon Bedrock's multi-region inference automatically doubles capacity by load balancing across regions  * Start small with managed services and scale resiliency practices with system maturity  Tags  #GenAI #EdTech #AWS #Resiliency #AmazonBedrock #HigherEducation #AIinEducation #CloudArchitecture #DigitalTransformation #StudentSuccess

S1E23: 23: CoTeacher - AI-Powered Support for Inclusive Education

Episode Details  * Date: March 30, 2026  * Duration: ~23 minutes  * Host: Pranusha Manchala   * Guests:   * Darren Kraker, Solutions Architect, AWS Cloud Innovation Center  * Ryan Maine, Director of Student Services, Walnut Valley Unified School District  * Jeremy Goins, Deputy Superintendent, Corona Norco Unified School District  * Sharon Liang, Software Engineer Intern Episode Summary  This episode explores CoTeacher, an innovative generative AI-powered tool developed through collaboration between AWS Cloud Innovation Center and California school districts. The solution addresses a critical challenge in education: helping teachers efficiently manage and implement individualized accommodations for diverse learners. CoTeacher instantly surfaces required student accommodations from IEPs and 504 plans while providing actionable teaching guidance, enabling educators to deliver more inclusive instruction confidently and efficiently.  Key Discussion Points  * The overwhelming demands on teachers managing diverse student needs, IEPs, and 504 plans  * How CoTeacher emerged from collaborative problem-solving between educators and AWS  * The rapid prototyping process at AWS Cloud Innovation Center (4-week MVP turnaround)  * Teacher involvement in product development and continuous feedback loops  * Building trust through data privacy, security, and responsible AI implementation  * AI as an accelerator for teachers, not a replacement  Featured Technologies  * Amazon Bedrock with Claude Sonnet model  * AWS Lambda  * Amazon API Gateway  * Amazon DynamoDB  * Amazon Cognito  * AWS Amplify  Key Takeaways  * Teachers manage 180+ students with varying IEPs, 504s, and English language learner needs  * CoTeacher provides a "scouting report" on each student, surfacing strengths, challenges, and proven strategies  * The solution uses serverless, API-driven architecture for security and scalability  * Teachers remain in the driver's seat while AI accelerates their effectiveness  * The tool enables seamless knowledge transfer as students progress through grades  Resources Mentioned  * AWS Cloud Innovation Center  * IEPs (Individual Education Plans)  * 504 Plans  * ELPAC Testing  Tags  #EdTech #InclusiveEducation #AWS #GenerativeAI #TeacherSupport #SpecialEducation #K12Education #AmazonBedrock #CloudInnovation #StudentSuccess #AIinEducation #DigitalTransformation

S1E22: 22: Math Fact Lab and its scale on AWS

Episode Details  * Date: March 16, 2026  * Duration: ~26 minutes  * Host: Pranusha Manchala   * Guest: Mike Kenny, Owner of Math Fact Lab  Episode Summary  In this episode, Mike Kenny shares the inspiring journey of Math Fact Lab, a platform revolutionizing how students master foundational math skills. As a former fifth-grade math teacher with 19 years of experience, Mike discusses how he transformed his classroom materials into a cloud-hosted application that helps students develop math fact fluency through research-based strategies rather than traditional memorization. He reveals how AWS has supported Math Fact Lab's growth from a bootstrapped startup to serving schools and districts nationwide, emphasizing the importance of conceptual understanding and personalized learning in mathematics education.  Key Discussion Points  * Evolution from classroom teacher to EdTech entrepreneur  * Research-based approach to math fact fluency vs. traditional memorization  * Platform design for personalized learning with customizable settings  * Scaling from individual teachers to school districts with rostering integrations  * Student success metrics showing 40% to 90% improvement in fact mastery  * AWS's role in enabling secure, scalable infrastructure for student data  * Future expansion into integer operations for middle school students  Featured Technologies  * AWS Cloud Infrastructure  * Clever and ClassLink rostering integrations  * Visual learning models (dice, area models, ten frames)  * Adaptive assessment systems  Key Takeaways  * Strategy-based learning outperforms memorization for long-term math fluency  * Personalization (adjustable time limits, learning tracks) accommodates diverse student needs  * AWS startup programs provided critical early-stage support for bootstrapped ventures  * Customer feedback drives continuous product evolution and feature development  Resources Mentioned  * Math Fact Lab: mathfactlab.com  * AWS Startup Programs  * Research on math anxiety and timed tests  Tags  #EdTech #MathEducation #AWS #StudentSuccess #K12Education #DigitalLearning #CloudComputing #StartupJourney #PersonalizedLearning #Education