50+

Sessions
Ask the
Experts
Live Q&A
Customer
Stories
Use Cases
ML
Experience
Learn more
Key
Concepts
Innovation
logo-intel-2021v2

50+

Sessions
Ask the
Experts
Live Q&A
Customer
Stories
Use Cases
ML
Experience
Learn more
Key
Concepts
Innovation
logo-intel-2021v2

Agenda

Get inspired and learn how you can use machine learning to drive better experiences, streamline operations, and reduce risks, and walk away with the ability to implement these projects for your organization. Dive deep into any of the 50+ business and technical sessions led by AWS experts as they share the latest innovations in AI/ML, key concepts, business use cases, architectural best practices, and answer your questions live.

Use the breaks to visit our Innovation Lab, Partner Pavillion, DeepRacer Zone and connect with an AWS Expert to dive deeper into our AI/ML content.

 Download Agenda at a Glance »

Sessions

  • Opening Keynote
  • I’m an Application Developer
  • I’m a Data Scientist
  • I’m a Data Engineer
  • I’m an MLOps Engineer
  • I’m a Technical Decision Maker
  • I’m a Business Decision Maker
  • Closing Keynote
  •  German
  •  French
  •  Spanish
  •  Hebrew
  •  Italian
  •  Russian
  •  Polish
  • Opening Keynote
  • Machine learning is one of the most disruptive technologies we will encounter in our generation. More than 100,000 customers use AWS for machine learning today, in all industry segments. In this Keynote, we’ll give you an overview of the services that these customers are using to innovate faster and improve productivity. We’ll also introduce the different tracks of the conference to help you find the best one and make the most of your day.

    Speaker: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS
    Duration: 30 minutes

  • I’m an Application Developer
  • About the track

    Learn how to quickly and easily add AI capabilities and ML models to your applications. No need for expert skills: just call APIs or use off the shelf models, and get the job done!

    Add your own computer vision model to your application without ML skills (Level 200)

    Computer vision allows machines to identify people, places, and things in images with accuracy at or above human levels with much greater speed and efficiency. The applications of computer vision are far-reaching, from identifying defects in high speed assembly lines to the analysis of medical images.

    In this talk we will build an automated content moderation system for your application using Amazon Rekognition. The demo will flag potentially unsafe or inappropriate content across both image and video assets with detailed labels that allow you to accurately control what you want to allow. The best part? The use of Rekognition requires no prior machine learning expertise to use.

    Speaker: Sohan Maheshwar, Developer Advocate, AWS
    Duration: 30 minutes


    Build your own defect and anomaly detection models without ML skills (Level 300)

    How can development teams add smart capabilities to business applications without any ML skills? In this hands-on session we will focus on two specific use cases related to anomaly detection. We'll dive into the practical steps of how to identify product defects in images and how to detect outliers and issues in business metrics.

    Speaker: Alex Casalboni, Senior Developer Advocate, AWS
    Duration: 30 minutes


    Deploy state of the art ML models and solutions in a single click (Level 300)

    There are many challenges in getting started with machine learning (ML). Developers who are new to this field often struggle with importing a model from a popular model zoo and deploying it to an API endpoint. Additional components are also needed to launch a working ML application, including API gateways, serverless compute, object storage, ETL streams, dashboards, and authentication. Thus, the end-to-end process of building a solution can take months or longer for new ML users. In this talk, I will show you can easily and quickly bring ML applications to market. we will demonstrate how to deploy solutions using out-of-the-box ML models, and how to customize them for a specific business problem. With just a few clicks, we will show you how to launch ML solutions preconfigured with all AWS resources required for production, including a CloudFormation template and a reference architecture.

    Speaker: Sébastien Stormacq, Principal Developer Advocate, AWS
    Duration: 30 minutes


    Expert corner: Let’s chat with Luca Bianchi, AWS Serverless Hero (Level 200)

    In this session, Marcia will chat with Luca Bianchi, an AWS Serverless Hero and CTO at Neosperience. We will be discussing if AI managed services can produce the same results as custom models trained with Sagemaker. We will also be giving some tips for developers that are working in the ML world and sharing as many best practices as we can fit in 30 minutes.

    Speakers: Marcia Villalba, Senior Developer Advocate, AWS | Luca Bianchi, AWS Serverless Hero
    Duration: 30 minutes

  • I’m a Data Scientist
  • About the track

    Learn how to build, train, and deploy high-quality models at any scale, without having to worry about infrastructure. Focus on understanding the ML problem at hand, and on how to solve it using either existing algorithms or your own.

    Move and scale your ML experiments in the cloud (Level 200)

    If you have been running your notebooks and machine learning experiments on your laptop or fighting for resources with your colleagues on some common Jupyter server, struggling to share and collaborate without tracking dependencies and other restrictions, it’s time to scale your ML environment to the cloud.

    This session will introduce you to Amazon SageMaker and especially SageMaker notebooks, will allow you to access notebooks in seconds without spinning up compute instances, easily scale compute up or down, share notebooks, so others can reproduce the results with the same data, environment and library dependencies, access other AWS resources and control your security and allow you to build models faster and collaborate at scale.

    Speaker: Boaz Ziniman, Principal Developer Advocate, AWS
    Duration: 30 minutes


    Detect potential bias in your datasets and explain how your models predict (Level 300)

    As ML models are built by training algorithms that learn statistical patterns present in datasets, several questions immediately come to mind. First, can we ever hope to explain why our ML model comes up with a particular prediction? Second, what if our dataset doesn’t faithfully describe the real-life problem we were trying to model? Could we even detect such issues? Would they introduce some sort of bias in imperceptible ways? These are not speculative questions at all, and their implications can be far-reaching. Unfortunately, even with the best of intentions, bias issues may exist in datasets and be introduced into models with business, ethical, and legal consequences. 

    It is thus important for model builders and administrators to be aware of potential sources of bias in production systems. In addition, many companies and organizations need ML models to be explainable before they can be used in production. In fact, some regulations explicitly require model explainability for consequential decision making. In this hands-on session, you’ll learn how Amazon SageMaker Clarify can help you tackle bias and explainability issues, and how to use it with both the SageMaker Studio user interface and the SageMaker SDK. You’ll also see how it works together with SageMaker Model Monitor to track bias metrics over time on your prediction endpoints.

    Speaker: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS
    Duration: 30 minutes


    Expert corner: Let’s chat with Francesco Pochetti, AWS Machine Learning Hero (Level 300)

    In this session, Julien will chat with Francesco Pochetti, an AWS Machine Learning Hero and a seasoned Data Scientist. Discussing the concrete tasks that data scientists work on daily, as well as new requirements such as fairness and explainability, they'll share as many best practices and tips that they can fit in 30 minutes, in order to help you build high-quality models faster.

    Speakers: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS | Francesco Pochetti, AWS Machine Learning Hero
    Duration: 30 minutes


    Scale your large training jobs with data and model parallelism (Level 300)

    Advances in deep learning have led to use cases such as computer vision and natural language processing models, where training time and model size can create a bottleneck. To speed up training, developers can use data parallelism to train the model in parallel on shards of data on a cluster of GPU instances. To train models that are memory-constrained and don't fit in a single GPU's memory, developers can use model parallelism to partition partition model across multiple GPUs. However, efficiently scaling to multiple GPU instances and implementing model parallelism can be very difficult and time consuming requires deep expertise in deep learning architectures, distributed systems, and software frameworks. In this session learn how to use the new distributed DataParallel and ModelParallel library in Amazon SageMaker to reduce training time and costs, and easily apply model parallelism to scale to billions of parameters using both TensorFlow and PyTorch while maintaining high scaling efficiency.

    Speaker: Shashank Prasanna, Senior Developer Advocate, AI/ML, AWS
    Duration: 30 minutes

  • I’m a Data Engineer
  • About the track

    Learn how to turn raw data into clean and expressive ML datasets, without having to manage any infrastructure. Focus on exploring and enriching data coming from a variety of sources, in order to build high-quality models.

    Manage your ML data in a central repository (Level 300)

    Before you start wrangling data and creating features, you need to have a way to move the raw data from multiple data stores into a central repository you can use as a starting point. In this session you will see how data can be ingested out of SQL, NoSQL, or even streaming datasources, and how you can then explore and combine the data to build ML datasets using either SQL or visual data preparation tools.

    Speaker: Javier Ramirez, Developer Advocate, AWS
    Duration: 30 minutes


    Expert corner: Let’s chat with Walter Riviera, AI Technical Specialist at Intel (Level 300)

    In this session, Julien will chat with Walter Riviera, an AI Technical Specialist at Intel. They'll share technical insights on how to best use Intel solutions to run your ML workloads on AWS. Along the way, they'll also discuss Intel hardware and software technology such as Intel Skylake, AVX512, DLBoost, the Math Kernel Library, OpenVino, and more. Of course, they'll also tell you more about the newly-announced EC2 Habana Gaudi instances, which will be available on AWS in 2021. Get ready to learn a lot!

    Speakers: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS | Walter Riviera, Data Scientist, Intel
    Duration: 30 minutes


    Prepare your datasets at scale using Apache Spark and SageMaker Data Wrangler (Level 300)

    Pandas doesn’t scale for large datasets. Apache Spark is an open source, distributed processing engine that scales to large datasets across a large number of cluster instances. In this session, I will demonstrate several ways to use Apache Spark on AWS to analyze large datasets, perform data quality and bias checks, transform raw data into machine learning features, and train predictive models.

    Speaker: Chris Fregly, Senior Developer Advocate, AI/ML, AWS
    Duration: 30 minutes


    Standardize and automate your feature engineering workflows (Level 300)

    As a data scientist, you certainly spend a lot of time crafting feature engineering code. Indeed, given the experimental nature of this work, even a small project can lead to multiple iterations. Thus, you’ll often run the same feature engineering code again and again, wasting time and compute resources on repeating the same operations. In large organizations, this may cause an even greater loss of productivity, as different teams often run identical jobs, or even write duplicate feature engineering code because they have no knowledge of prior work. 

    As models are trained on engineered datasets, it’s also imperative that you apply the same transformations to data used for prediction. This often means rewriting your feature engineering code (sometimes in a different language), integrating it in your prediction workflow, and running it at prediction time. This whole process is not only time-consuming, it can also introduce inconsistencies, as even the tiniest variation in your data transforms can have a large impact on predictions. In this hands-on session, you’ll learn how to solve all these problems with Amazon SageMaker Feature Store, and how to use it with both the SageMaker Studio user interface and the SageMaker SDK. You’ll also see how it works together with SageMaker Data Wrangler to simplify your end to end data preparation workflows.

    Speaker: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS
    Duration: 30 minutes

  • I’m an MLOps Engineer
  • About the track

    Learn how to support data science and data engineering teams in the most efficient way. Automate workflows end-to-end thanks to AWS and open source tools, and select the best infrastructure for each use case.

    Expert corner: Let’s chat with Pavlos Mitsoulis, AWS Machine Learning Hero (Level 200)

    In this session, Cobus Bernard will chat with Pavlos Mitsoulis, a Staff Data Scientist from Expedia Group and AWS Machine Learning Hero, about what MLOps is, how it is different from DevOps and the typical challenges companies face when trying to automate their ML workflows.

    Speakers: Cobus Bernard, Senior Developer Advocate, AWS | Pavlos Mitsoulis, AWS Machine Learning Hero
    Duration: 30 minutes


    Select the right ML instance for your training and inference jobs (Level 300)

    AWS offers a breadth and depth of Machine Learning (ML) infrastructure for training and inference workloads that you can use through either a do-it-yourself approach or a fully managed approach with Amazon SageMaker. In this session, explore how to choose the proper instance for ML training and inference based on model size, complexity, throughput, framework choice, inference latency and portability requirements. Join this session to compare and contrast compute-optimized CPU-only instances, such as Amazon EC2 C4 and C5; high-performance GPU instances, such as Amazon EC2 G4, P3, and P4d; cost-effective variable-size GPU acceleration with Amazon Elastic Inference; and high performance/cost with Amazon EC2 Inf1 instances powered by custom-designed AWS Inferentia chips.

    Speaker: Shashank Prasanna, Senior Developer Advocate, AI/ML, AWS
    Duration: 30 minutes


    Automate your ML workflows with end-to-end pipelines (Level 300)

    Developing a high-quality ML model involves many steps. We typically start with exploring and preparing our data. We experiment with different algorithms and parameters. We spend time training and tuning our model until the model meets our quality metrics, and is ready to be deployed into production. Orchestrating and automating workflows across each step of this model development process can take months of coding.

    In this session, we show you how to create, automate, and manage end-to-end ML workflows using Amazon SageMaker Pipelines. We will create a reusable NLP model pipeline to prepare data, store the features in a feature store, fine-tune a BERT model, and deploy the model into production if it passes our defined quality metrics.

    Speaker: Antje Barth, Senior Developer Advocate for AI and Machine Learning, AWS
    Duration: 30 minutes


    Build your ML platforms using open source technology (Level 300)

    As businesses adopt machine learning, they are looking to automate and orchestrate the various stages of the machine learning pipeline. Whether it is data engineers orchestrating the various stages of data acquisition and preparation, helping data scientists be able to track and manage local development of models, or how you deploy the finished models to production, many look to use open source tools. There are many tools to choose from and in this session we are going to be exploring a few of these.

    Speaker: Ricardo Sueiras, Principal Developer Advocate, AWS
    Duration: 30 minutes

  • I’m a Technical Decision Maker
  • About the track

    There’s more to ML than training and deploying models, especially in large organizations. Learn about AWS best practices for AI/ML, which will help you make your ML workflows more efficient, resilient, and secure.

    Break down data silos: Build a serverless data lake on Amazon S3 (Level 200)

    Flexibility, security, performance, and optimizing costs are key when building and scaling a data lake. The analytics solutions you use in the future will almost certainly be different from the ones you use today, and choosing the right storage foundation gives you the agility to quickly experiment and migrate with the latest analytics solutions. In this session, explore the best practices for optimizing your storage, performance, and costs when building a data lake in Amazon S3 and Amazon S3 Glacier.

    Duration: 
    30 minutes


    From POC to production: Strategies for achieving machine learning at scale (Level 200)

    You’ve put your data strategy in place, found the right use case, and successfully implemented your first proof of concept (POC). Now what? The key to machine learning success is scale. Yet, many organizations run into challenges when attempting to move machine learning models from an initial POC to production and to true organizational scale. In this session, executives and managers who are looking to achieve success using machine learning at scale within their organizations can get guidance, including best practices for MLOps, data governance, and knowledge sharing.

    Duration: 30 minutes


    Architectural best practices for machine learning applications (Level 300)

    Machine learning (ML) application architecture is both the same and different from other architectures deployed in the AWS Cloud. As cloud-based ML applications gain in popularity, users are increasingly asking, “Am I building this right?” This session covers how the best practices established in the AWS Well-Architected Framework apply to ML workloads and where distinctions need to be made to address the unique nature of ML architecture. As a case study, the session also dives deep on publicly available AWS Well-Architected solutions to help you ensure that you have the tools you need to get started correctly.

    Duration: 30 minutes


    Secure and compliant machine learning for regulated industries (Level 300)

    As organizations bring their machine learning (ML) workloads to the cloud, having access to an environment that is secure is the top requirement. In this session, learn the steps involved in provisioning a secure ML environment on Amazon SageMaker, dive deep into common customer patterns and architectures, and see how to leverage other AWS services to build these environments in a consistent and reproducible manner.

    Duration: 30 minutes

  • I’m a Business Decision Maker
  • About the track

    For all its technical complexity, the main appeal of AI/ML is how it helps organizations innovate faster and improve their business outcomes. Learn from companies that have successfully done it, and how you can do the same!

    Innovation is never normal (Level 100)

    In 2020 the phrase ‘Never Normal’ became common language. And like most periods of major upheaval, the first instinct of some leaders is to focus on survival. For businesses working with AI and ML however, living this never normal is simply ‘business as usual’, where constant change offers abundant opportunities to innovate, and thrive. Join Olivier Klein, Lead Architect AWS as he presents customer stories.

    Duration: 45 minutes


    Amazon.com’s use of AI/ML to enhance the customer experience (Level 100)

    Amazon.com uses AI/ML in innovative and scaled ways to transform the way we operate and invent new customer experiences. In this session, targeted at senior business and technology decision makers, we share specific examples from Amazon.com’s consumer/retail and other businesses to explain how AI/ML helps Amazon deliver the best customer experience possible while improving efficiency and lowering cost. We cover the insights and lessons Amazon.com learned across the cultural, process, and technology aspects of building and scaling ML capabilities in the organization.

    Duration: 30 minutes


    How do you innovate to drive business outcomes? (Level 200)

    In this session, hear real-life stories of how customers have used AWS Cloud and AWS Professional Services to accelerate technical innovation and drive transformational business outcomes. Hear examples across industries, use cases, and technology stacks.

    Duration: 30 minutes


    From POC to production: Strategies for achieving machine learning at scale (Level 200)

    You’ve put your data strategy in place, found the right use case, and successfully implemented your first proof of concept (POC). Now what? The key to machine learning success is scale. Yet, many organizations run into challenges when attempting to move machine learning models from an initial POC to production and to true organizational scale. In this session, executives and managers who are looking to achieve success using machine learning at scale within their organizations can get guidance, including best practices for MLOps, data governance, and knowledge sharing.

    Duration: 30 minutes

  • Closing Keynote
  • In the closing keynote, we’ll focus on AWS resources that will help you kickstart and accelerate your machine learning projects on AWS: customer use cases, partners, whitepapers, reference architectures, solutions, blogs, and code! 10 minutes packed with real-life and actionable content, so don’t miss it!

    Speaker: Julien Simon, Principal Developer Advocate, AI & Machine Learning, AWS
    Duration: 10 minutes

  •  German
  • Opening Keynote

    Speaker: Antje Barth, Senior Developer Advocate for AI and Machine Learning, AWS
    Duration: 30 minutes


    Erstelle eigene ML-Modelle zur Fehler- und Anomalie-Erkennung – sogar ohne Vorkenntnisse in ML (Level 300)

    Speaker: Stefan Christoph, Sr. Solutions Architect, AWS
    Duration: 30 minutes


    Fairness in der KI: Voreingenommenheit in Datensätzen erkennen und Modellvorhersagen besser verstehen (Level 300)

    Speaker: Johannes Langer, Sr. Solutions Architect, AWS
    Duration: 30 minutes


    Automatisiere ML Workflows mit Ende-zu-Ende Pipelines (Level 300)

    Speaker: Antje Barth, Senior Developer Advocate for AI and Machine Learning, AWS
    Duration: 30 minutes


    Kundenvortrag: Bundesliga Match Facts Powered by AWS (Level 200)

    Speaker: Marco Salazar, Application Developer, AWS Professional Services | Mirko Janetzke, Director Product & Analytics, Sportec Solutions AG
    Duration: 30 minutes

  •  French
  • Keynote d’ouverture

    Speaker: Sébastien Stormacq, Principal Developer Advocate, AWS
    Duration: 30 minutes


    Déployez des modèles et des solutions ML à la pointe, en un seul clic (Niveau 300)

    Speaker: Sébastien Stormacq, Principal Developer Advocate, AWS
    Duration: 30 minutes


    Construire votre propre modèle de détection de défauts et d’anomalies sans compétence de machine learning (Niveau 300)

    Speaker: Michaël Hoarau, Senior AI/ML Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Sélectionner la bonne instance de machine learning (Niveau 200)

    Speaker: Davide Gallitelli, Solutions Architect, AWS
    Duration: 30 minutes


    Expliquer les prédictions de vos modèles et en détecter les biais potentiels (Niveau 300)

    Speaker: Segolene Dessertine-Panhard, Sr. Data Scientist, AWS
    Duration: 30 minutes

  •  Spanish
  • Keynote de apertura

    Speaker: Javier Ramirez, Developer Advocate, AWS
    Duration: 30 minutes


    Crea tus propios modelos para detectar defectos y anomalías sin ser experto en aprendizaje automático (Nivel 300)

    Speaker: Anna Grüebler, Senior AI Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Estandariza y automatiza tus procesos de ingeniería de características (Nivel 300)

    Speaker: Javier Ramirez, Developer Advocate, AWS
    Duration: 30 minutes


    Identifica posible sesgo en tus conjuntos de datos y explica cómo funcionan tus modelos (Nivel 300)

    Speaker: María Gaska, Senior AI/ML Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Automatiza tus procesos de ML con pipelines de extremo a extremo (Nivel 300)

    Speaker: Antonio Rodriguez, Senior AI/ML Specialist Solutions Architect, AWS
    Duration: 30 minutes

  •  Hebrew
  • Opening Keynote

    Speaker: Boaz Ziniman, Principal Developer Advocate, Israel, AWS
    Duration: 30 minutes


    Scale your large training jobs with data and model parallelism (Level 400)

    Speaker: Gili Nachum, Senior AI/ML Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Manage your ML data in a central repository (Level 300)

    Speaker: Orit Alul, Principal Solutions Architect, AWS
    Duration: 30 minutes


    Standardize and automate your feature engineering workflows (Level 300)

    Speaker: Gili Nachum, Senior AI/ML Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Detect potential bias in your datasets and explain how your models predict (Level 300)

    Speaker: Eitan Sela, Sr Startup Solutions Architect, AWS
    Duration: 30 minutes

  •  Italian
  • Keynote di apertura

    Speaker: Alex Casalboni, Sr. Developer Advocate, AWS
    Duration: 30 minutes


    Crea e utilizza modelli per il rilevamento di anomalie senza competenze di ML (Livello 300)

    Speaker: Woody Borraccino, Sr Specialist Solutions Architect, AWS
    Duration: 30 minutes


    Come standardizzare e automatizzare il processo di feature engineering (Livello 300)

    Speaker: Veziona Ekonomi, Sr Solutions Architect, AWS
    Duration: 30 minutes


    ML sul campo: quattro chiacchiere con Luca Bianchi, AWS Serverless Hero (Livello 200)

    Speaker: Alex Casalboni, Sr. Developer Advocate, AWS | Luca Bianchi, AWS Serverless Hero
    Duration: 30 minutes


    Come automatizzare i processi di ML con una pipeline end-to-end (Livello 300)

    Speaker: Giuseppe Angelo Porcelli, Principal ML Specialist Solutions Architect, AWS
    Duration: 30 minutes

  •  Russian
  • Opening Keynote

    Speaker: Александр Ложечкин (Alexander Lozhechkin), Senior Manager, Partner Solutions Architecture EMEA, AWS
    Duration: 30 minutes


    Добавьте компьютерное зрение в ваше приложение без навыков машинного обучения (Level 200)

    Speaker: Михаил Голубев (Mike Golubev), Solutions Architect, AWS
    Duration: 30 minutes


    Создавайте модели обнаружения дефектов и аномалий даже не имея навыков в машинном обучении (Level 200)

    Speaker: Александр Изюмов (Aleksandr Iziumov), Solutions Architect, AWS
    Duration: 30 minutes


    Управляйте вашими ML данными в одном централизованном и безопасном репозитории (Level 300)

    Speaker: Егор Шадрин (Egor Shadryn), Solutions Architect Startups, AWS
    Duration: 30 minutes


    Подготовьте ваши наборы данных - обрабатывайте данные любого объема с помощью Apache Spark и SageMaker Data (Level 300)

    Speaker: Игорь Иванюк (Igor Ivaniuk), Solutions Architect, AWS
    Duration: 30 minutes

  •  Polish
  • Opening Keynote

    Speaker: Tomasz Stachlewski, Senior Solutions Architecture Manager, AWS
    Duration: 30 minutes

Session levels designed for you

Foundational
Level 100

Sessions are focused on providing an overview of AWS services and features, with the assumption that attendees are new to the topic.

INTERMEDIATE
Level 200

Sessions are focused on providing best practices, details of service features and demos with the assumption that attendees have introductory knowledge of the topics.

ADVANCED
Level 300

Sessions dive deeper into the selected topic. Presenters assume that the audience has some familiarity with the topic, but may or may not have direct experience implementing a similar solution.

EXPERT
Level 400

Sessions are for attendees who are deeply familiar with the topic, have implemented a solution on their own already, and are comfortable with how the technology works across multiple services, architectures, and implementations.

Get started with AWS today

Gain free, hands-on experience with the AWS platform, products, and services. Explore more than 60 products and start building on AWS using the free tier.