As generative AI continues to transform how enterprises operate—and develop net new innovations—the infrastructure demands for training and deploying AI models have grown exponentially. Traditional infrastructure approaches are struggling to keep pace with today’s computational requirements, network demands, and resilience needs of modern AI workloads.

At AWS, we’re also seeing a transformation across the technology landscape as organizations move from experimental AI projects to production deployments at scale. This shift demands infrastructure that can deliver unprecedented performance while maintaining security, reliability, and cost-effectiveness. That’s why we’ve made significant investments in networking innovations, specialized compute resources, and resilient infrastructure that’s designed specifically for AI workloads.

Accelerating model experimentation and training with SageMaker AI

The gateway to our AI infrastructure strategy is Amazon SageMaker AI, which provides purpose-built tools and workflows to streamline experimentation and accelerate the end-to-end model development lifecycle. One of our key innovations in this area is Amazon SageMaker HyperPod, which removes the undifferentiated heavy lifting involved in building and optimizing AI infrastructure.

At its core, SageMaker HyperPod represents a paradigm shift by moving beyond the traditional emphasis on raw computational power toward intelligent and adaptive resource management. It comes with advanced resiliency capabilities so that clusters can automatically recover from model training failures across the full stack, while automatically splitting training workloads across thousands of accelerators for parallel processing.

The impact of infrastructure reliability on training efficiency is significant. On a 16,000-chip cluster, for instance, every 0.1% decrease in daily node failure rate improves cluster productivity by 4.2% —translating to potential savings of up to $200,000 per day for a 16,000 H100 GPU cluster. To address this challenge, we recently introduced Managed Tiered Checkpointing in HyperPod, leveraging CPU memory for high-performance checkpoint storage with automatic data replication. This innovation helps deliver faster recovery times and is a cost-effective solution compared to traditional disk-based approaches.

For those working with today’s most popular models, HyperPod also offers over 30 curated model training recipes, including support for OpenAI GPT-OSS, DeepSeek R1, Llama, Mistral, and Mixtral. These recipes automate key steps like loading training datasets, applying distributed training techniques, and configuring systems for checkpointing and recovery from infrastructure failures. And with support for popular tools like Jupyter, vLLM, LangChain, and MLflow, you can manage containerized apps and scale clusters dynamically as you scale your foundation model training and inference workloads.

Overcoming the bottleneck: Network performance

As organizations scale their AI initiatives from proof of concept to production, network performance often becomes the critical bottleneck that can make or break success. This is particularly true when training large language models, where even minor network delays can add days or weeks to training time and significantly increase costs. In 2024, the scale of our networking investments was unprecedented; we installed over 3 million network links to support our latest AI network fabric, or 10p10u infrastructure. Supporting more than 20,000 GPUs while delivering 10s of petabits of bandwidth with under 10 microseconds of latency between servers, this infrastructure enables organizations to train massive models that were previously impractical or impossibly expensive. To put this in perspective: what used to take weeks can now be accomplished in days, allowing companies to iterate faster and bring AI innovations to customers sooner.

At the heart of this network architecture is our revolutionary Scalable Intent Driven Routing (SIDR) protocol and Elastic Fabric Adapter (EFA). SIDR acts as an intelligent traffic control system that can instantly reroute data when it detects network congestion or failures, responding in under one second—ten times faster than traditional distributed networking approaches.

Accelerated computing for AI

The computational demands of modern AI workloads are pushing traditional infrastructure to its limits. Whether you’re fine-tuning a foundation model for your specific use case or training a model from scratch, having the right compute infrastructure isn’t just about raw power—it’s about having the flexibility to choose the most cost-effective and efficient solution for your specific needs.

AWS offers the industry’s broadest selection of accelerated computing options, anchored by both our long-standing partnership with NVIDIA and our custom-built AWS Trainium chips. This year’s launch of P6 instances featuring NVIDIA Blackwell chips demonstrates our continued commitment to bringing the latest GPU technology to our customers. The P6-B200 instances provide 8 NVIDIA Blackwell GPUs with 1.4 TB of high bandwidth GPU memory and up to 3.2 Tbps of EFAv4 networking. In preliminary testing, customers like JetBrains have already seen greater than 85% faster training times on P6-B200 over H200-based P5en instances across their ML pipelines.

To make AI more affordable and accessible, we also developed AWS Trainium, our custom AI chip designed specifically for ML workloads. Using a unique systolic array architecture, Trainium creates efficient computing pipelines that reduce memory bandwidth demands. To simplify access to this infrastructure, EC2 Capacity Blocks for ML also enable you to reserve accelerated compute instances within EC2 UltraClusters for up to six months, giving customers predictable access to the accelerated compute they need.

Preparing for tomorrow’s innovations, today

As AI continues to transform every aspect of our lives, one thing is clear: AI is only as good as the foundation upon which it is built. At AWS, we’re committed to being that foundation, delivering the security, resilience, and continuous innovation needed for the next generation of AI breakthroughs. From our revolutionary 10p10u network fabric to custom Trainium chips, from P6e-GB200 UltraServers to SageMaker HyperPod’s advanced resilience capabilities, we’re enabling organizations of all sizes to push the boundaries of what’s possible with AI. We’re excited to see what our customers will build next on AWS.

About the author

Barry Cooks is a global enterprise technology veteran with 25 years of experience leading teams in cloud computing, hardware design, application microservices, artificial intelligence, and more. As VP of Technology at Amazon, he is responsible for compute abstractions (containers, serverless, VMware, micro-VMs), quantum experimentation, high performance computing, and AI training. He oversees key AWS services including AWS Lambda, Amazon Elastic Container Service, Amazon Elastic Kubernetes Service, and Amazon SageMaker. Barry also leads responsible AI initiatives across AWS, promoting the safe and ethical development of AI as a force for good. Prior to joining Amazon in 2022, Barry served as CTO at DigitalOcean, where he guided the organization through its successful IPO. His career also includes leadership roles at VMware and Sun Microsystems. Barry holds a BS in Computer Science from Purdue University and an MS in Computer Science from the University of Oregon.

By Patrick Supanc, Chief Product Officer

Today at Coursera Connect, our annual conference, we announced the launch of Skill Tracks, our data-backed learning solution mapped to specific occupations that guides learners from foundational knowledge to expert proficiency through verified skill paths. 

Skill Tracks are powered by Coursera’s Career Graph, our proprietary system that analyzes millions of labor market data points, third-party competency frameworks, and our skills taxonomy, to precisely map the relationships between jobs, skills, and learning content, ensuring organizations can close skill gaps quickly.

View a Skill Tracks video here.

The World Economic Forum’s Future of Jobs Report 2025 finds that 63% of employers see skill gaps as the biggest barrier to business transformation, with nearly 6 in 10 workers needing reskilling within the next five years. With Coursera Skill Tracks, leaders can ensure their teams have the right skills to boost innovation, productivity, and retention.

Key features include:

• A tailored learning experience – In addition to world-class content from industry leaders and universities like Microsoft, AWS, Yale, and Stanford, learning leaders can customize Skill Tracks with their own content, ensuring alignment with their organization’s specific tools, workflows, and business priorities.
• Rigorous and verifiable credentials – Learners progress toward credentials based on real-world assessments, providing both motivation and proof that skills are not only learned but also demonstrated.
• Real-time insights and alignment to business goals – Regular tracking of learning progress and continuous content updates ensure Skill Tracks stay current with market demands, align with changing skill requirements for roles, and directly connect skill acquisition with business performance and growth.

Starting today, four Skill Tracks are available:

• Software and Product – Covers the most critical skills in mobile development, product management, UX design, web development, and software development 
• IT – Includes necessary skills in computer systems and architecture, cybersecurity, IT management, IT support and operations, and network engineering 
• Data – Develops critical skills in data analysis, data engineering, data management, and AI and machine learning
• GenAI – Teaches practical applications of artificial intelligence for employees and leaders across customer service, human resources, data, finance, legal, marketing, product, sales, and more

Over the coming months, we’ll introduce additional Skill Tracks and enhanced features, including skill diagnostics to help learners start at the right level and verified skill paths with performance-based skills evaluation to produce credentials that reflect practical, job-ready expertise.

According to Matthew Dearmon, Ph.D, Informatica’s Senior Director, Talent Management and Leadership Development, “At Informatica, we have the only data management platform powered end-to-end by artificial intelligence, so it is vital that our teams and leaders are not just up-to-date, but are also looking ahead. Having a tailored learning solution aligned with real-time skills in demand for specific roles is essential to helping our technical leaders thrive when working with AI – and beyond.”

Paola Vera, Talent Management Head at Interbank added, “Coursera’s personalized learning approach has been a catalyst in helping Interbank build a data-driven, future-ready organization. Having targeted learning journeys informed by real-time labor data and aligned to specific occupations and career stages, can help ensure teams master the right skills for their role.”

Skill Tracks are available to existing Coursera customers with access to the full catalog. New customers can purchase Skill Tracks individually or bundled with the full catalog.  

Learn more about Coursera Skill Tracks and discover how a tailored, data-driven learning approach can accelerate skills development, technology adoption, or workforce transformation.

Click here to watch CEO Greg Hart’s keynote at Coursera Connect 2025, with a Skill Tracks demo and more.

By Marni Baker Stein, Chief Content Officer, Coursera

Today at Coursera Connect, our annual conference, we announced a major expansion of our partner network with several new world-class universities and forward-thinking industry leaders. Each new partnership deepens our commitment to helping learners around the world master the skills they need to grow their careers. 

In a world defined by rapid technological change, learners need flexible, affordable ways to keep pace. That’s why we’re expanding access to career-focused content across all levels of learning, from entry-level Professional Certificates to stackable courses that lead toward degrees. 

Welcoming our new partners
We’re proud to welcome our newest partner, Anthropic, one of the world’s leading AI research companies. Together, we’ll help learners and institutions apply the latest advances in AI, safely, effectively, and ethically — unlocking new ways to learn, teach, and work.

In addition to Anthropic, we’re also excited to welcome several new university and industry partners, expanding our reach across industries and disciplines:

• Hult International Business School – Global business education with campuses worldwide
• Minnesota State University, Mankato – Comprehensive U.S. public university
• University of the Arts London – Europe’s largest specialist creative arts university
• College of Engineering: University of Miami – Top rated Educational and Research University
• Universitat Politècnica de València – Spain’s leading STEM and engineering university
• UC Santa Barbara – Top 10 U.S. public research university

Leading organizations joining Coursera as industry partners include:

• AAPC – U.S. leader in medical billing and coding
• Harvard Business Publishing – Publisher of Harvard Business Review content
• ISSA (International Sports Sciences Association) – Global leader in fitness and wellness certification
• Pearson – Global leader in learning and assessments
• Skillshare – A creative learning community for personal and professional growth

These new partnerships strengthen our ability to deliver accessible, job-relevant learning for the modern workforce — spanning industries from healthcare and cybersecurity to generative AI, creative technology, and public policy.

Expanding Career Pathways with Courses and Professional Certificates
Professional Certificates on Coursera are designed to prepare learners for in-demand jobs, and many require no prior experience and can be completed in under six months. So far this year over 2.5M learners have enrolled in entry-level certs globally.

We’re excited to add new certificates from Microsoft, AAPC, and EC Council — creating even more opportunities for learners to gain job-ready skills.

• AAPC Medical Biller — Prepare for a career in healthcare billing by learning to process medical claims, manage reimbursements, and pass the Certified Professional Biller (CPB) exam.
• EC-Council Information Security Analyst — Learn to defend networks, investigate threats, and build job-ready cybersecurity skills in under five months.
• Coursera Python, SQL, Tableau for Data Science — Build practical skills to analyze, visualize, and present data insights. 
• Microsoft SQL Server — Learn to design, secure, and optimize databases with real-world projects for data careers.
• Microsoft R Programming for Everyone — Develop data analysis and visualization skills in R using Microsoft tools, GitHub Copilot, and Azure integration.
• Microsoft JavaScript Starter Kit — Master the fundamentals and frameworks to build interactive, portfolio-ready web applications.

Getting a head start on a degree with new AI and emerging tech courses from Illinois
The University of Illinois Urbana-Champaign, one of Coursera’s most innovative university partners, has launched four new open courses that explore emerging technologies shaping the future of work. These courses are stackable into Illinois’ for-credit iMBA and iMSA degree programs and offer learners a meaningful head start toward a flexible, affordable graduate degree. 

Courses now available:

These new offerings build on UIUC’s reputation for delivering high-quality, future-focused education on Coursera and offer learners an on-ramp to graduate-level learning with real career impact.

Coursera’s partner network includes some of the world’s most respected universities, industry leaders, and global organizations, all working together to deliver credentials that combine academic excellence with practical skill-building. As new technologies accelerate changes and lifelong learning becomes essential, we’re proud to work with our partners to expand access and create new opportunities for learners globally.

Watch Coursera CEO, Greg Hart’s keynote at Coursera Connect 2025 for demos of these new capabilities here.

Learn more about Skill Tracks here.

Read about our Product Announcements here.