<p><center><a href="https://www.datacamp.com/resources/webinars/operationalizing-data-within-large-organizations"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606930313/Maksim_webinar_CTA_y0xflz.png" /></a><br></p>
<p>What’s one thing that successful tech companies like Amazon, Netflix, Google, and Airbnb have in common? They’ve all scaled and operationalized their data science function. This means they’ve built top-notch data teams, and they’ve procured or built their own data tools. This required designing powerful internal processes and architectures that leverage data to move their business forward.</p>
<p>In our <a href="https://www.datacamp.com/resources/webinars/operationalizing-data-within-large-organizations">webinar last month</a>, Maksim Percherskiy, Data Engineer at The World Bank and former Chief Data Officer of the City of San Diego, explained that these companies were well positioned to do so because they grew up with data. “Older organizations—like financial institutions, healthcare institutions, and governments—didn&#39;t have the privilege [of building their business around data]. The city of San Diego was founded in the 1800s, and they&#39;re still learning to adjust and they&#39;re still learning how to effectively align data to their operations.”</p>
<p>Let’s unpack how large organizations can build an effective data science function. In recent years, many organizations have hired chief data officers, built teams of data scientists and data engineers, and found a way to monetize and share data. But NewVantage Partners’ <a href="http://newvantage.com/wp-content/uploads/2020/01/NewVantage-Partners-Big-Data-and-AI-Executive-Survey-2020-1.pdf">Big Data and AI Executive Survey</a> in January 2020 revealed that out of 70 firms surveyed, 98.8% had been investing in big data and AI initiatives, but only 37.8% were able to claim that they’d created a data-driven organization.</p>
<p>The most common barriers to becoming data-driven were people and process challenges—namely, operationalizing and building repeatable processes, creating a functional data infrastructure, and bridging the data skills gap.</p>
<h2 id="data-requirements">Data requirements</h2>
<p>The following visual summarizes what makes data usable and trustworthy:</p>
<p><center><a href="https://www.datacamp.com/resources/webinars/operationalizing-data-within-large-organizations"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1607609392/Ideal_data_world_-_1200x628_gz8mig.png" /></a><br></p>
<p>First, data must be <strong><em>collected</em></strong> at the right time, in the right manner, and with a purpose. Then, it must be <strong><em>discoverable</em></strong>, meaning all data users across the organization must know that this data has been collected and be able to find it. Next, the data must prove to be <strong><em>reliable</em></strong>, with no gaps or inconsistencies. It must also be easily <strong><em>understood</em></strong>—for instance, it should be correctly structured and labeled. Being <strong><em>compliant</em></strong> is critical—the proper security protocols must be in place to control access to sensitive data, and some industries have regulatory standards. And lastly, data must be <strong><em>actionable</em></strong>, meaning data users have the technology, training, and ethical frameworks to use data correctly.</p>
<h2 id="challenges-to-becoming-data-driven">Challenges to becoming data-driven</h2>
<p>There are three buckets of challenges to meeting these requirements of data: organizational, cultural, and technical.</p>
<h3 id="organizational-challenges">Organizational challenges</h3>
<p>Large organizations often have silos in which teams may have different objectives. These silos may be in place due to legal requirements, as is common in finance, or they may exist simply because teams may not have worked collaboratively in the past. Whatever the cause, silos cause miscommunication and misalignment, which harms the overall effectiveness of the organization.</p>
<p>A symptom of this could be managers who want to build bigger teams or the creation of more than one AI center of excellence.</p>
<h3 id="cultural-challenges">Cultural challenges</h3>
<p>Organizational data literacy—in which everyone has the data skills to be successful —is often thought of as a lofty goal, and some organizations may be risk averse to tackle it. This is why DataCamp tries to demystify topics like <a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders">machine learning</a> to help organizations understand how data can be of practical use. Empowering employees and making it easy to build skills is key.</p>
<p>Cultural challenges can also manifest in different incentives across the organization. Management may want eye-catching dashboards to present to their superiors, engineers want to use the latest and greatest tools, and business users may just want to answer a specific question or streamline processes. It’s not always possible to make every stakeholder happy. This ties into the organizational challenges mentioned previously—the goal is to align data strategy with overall business strategy.</p>
<h3 id="technical-challenges">Technical challenges</h3>
<p>Maksim says that large organizations often have “a patchwork of legacy systems” as the core systems of record for data. These systems can be difficult to pull data out of or scale—but often, department leads don’t want to take on a big project. They’re afraid of incurring the switching costs of replacing them, potential project failure costs, and additional retraining costs and a short-term hit to productivity. These are all shortsighted errors that result from a lack of unified data strategy.</p>
<p>Additional technical challenges include compliance and security standards, like HIPAA, FERPA, GDPR, and CCPA. Companies must adhere to these standards to avoid unethical data use.</p>
<p>Falling prey to these challenges results in what Brian Balfour describes as the <a href="https://brianbalfour.com/essays/growth-data-mistakes">Data Wheel of Death</a>, shown below. It illustrates that data that isn’t consistently maintained becomes irrelevant or flawed, causing people to lose trust in the data and ultimately, use it less.</p>
<p><center><a href="https://brianbalfour.com/essays/growth-data-mistakes"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1607609263/Data_wheel_-_1200x628_a9xrdn.png" /></a><br></p>
<h2 id="solutions-for-becoming-data-driven">Solutions for becoming data-driven</h2>
<p>So, how can well-established enterprises become data-driven like Netflix and Airbnb? They must rebuild around data. Here’s how.</p>
<h3 id="understand-the-landscape">Understand the landscape</h3>
<p>Large organizations must understand the different silos that exist and how they communicate with each other. They must cultivate champions at the executive level who will support this journey. And they must communicate the data objectives with all stakeholders.</p>
<h3 id="identify-your-users">Identify your users</h3>
<p>Next, to facilitate a successful data-driven transformation, you must <a href="https://www.datacamp.com/community/blog/eight-personas">understand your personas</a>. Empathizing deeply with each persona—whether they’re a data consumer, leader, data analyst, or data scientist—will help you to understand how best to help them and leverage them in your mission.</p>
<h3 id="start-small-and-keep-it-simple">Start small and keep it simple</h3>
<p>Maksim says that to begin, it’s not possible to fix everything at once. It’s really important to be disciplined about focusing on impact. Pick a project that is visible and impactful across the organization and that doesn’t require a ton of stakeholders or coordination.  Getting a series of easy wins to start with will catalyze buy-in and ensure the visibility of data projects. You’ll also be able to iterate on data work and leverage learnings for future data projects.</p>
<h3 id="align-data-strategy-with-business-strategy">Align data strategy with business strategy</h3>
<p>What does it mean to align data strategy with business strategy? Maksim’s practical advice is to choose data projects that expand and test your infrastructure, serve your business users, and align with company goals and results. Make sure every project is documented, reproducible, and follows best practices. Then, amplify your wins company-wide with executive support so you can build on those wins and keep iterating.</p>
<h2 id="what-being-data-driven-looks-like">What being data-driven looks like</h2>
<p>Being data-driven requires being able to measure the success of data initiatives. Maksim says it’s tempting to use easily accessible metrics like total number of projects, the number of completed datasets, or the number of repo followers.</p>
<p>“Those are all good metrics, but they don&#39;t really address the core of your success,” Maksim says. “I would think more about the number of decisions that get made with data. Also, how long it takes someone to find a set of data, or the time to get data. You should also look at how many goals, objectives, and OKRs are set and tracked with data, including how they’re reviewed.”</p>
<p>For more on making data operable within large organizations with complex processes, watch Maksim’s webinar on <a href="https://www.datacamp.com/resources/webinars/operationalizing-data-within-large-organizations">Operationalizing Data Within Large Organizations</a>.</p>
<p><center><a href="https://www.datacamp.com/resources/webinars/operationalizing-data-within-large-organizations"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606930313/Maksim_webinar_CTA_y0xflz.png" /></a><br></p>


https://www.datacamp.com/community/blog/data-driven-enterprise

One reason for the success of tech companies like Netflix and Airbnb: They’ve grown up with data. Not in tech? Here’s how to rebuild your company around data.

How Enterprise Organizations Can Become Data-Driven

<p>Python is the most popular programming language today and is widely used across verticals from software and web development, game development, data science, machine learning, and more. Learning Python is imperative for aspiring data scientists, data analysts, data engineers, and machine learning scientists.</p>
<p>The good news is that a complete beginner can learn and start programming in Python in just a couple of weeks. Here’s how to get started.</p>
<h2 id="know-your-motivation-before-you-begin">Know your motivation before you begin</h2>
<p>Python is a highly versatile programming language with many use cases. It’s important to have a specific motivation in mind as you embark on your Python learning journey. Targeting an area of interest will help you get started quicker and filter out learning resources that don’t fit your use case. My own Python learning journey began when I needed to use it for school for data science projects. I had no Python or programming experience and needed to get up to speed fast. Luckily, with a few online resources, I was able to become proficient in just a few months. Here’s a list of relevant areas where Python is helpful:</p>
<p><center><a href="https://datacamp.com"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606922669/Python_usecases_qfxctl.png" /></a> <br></p>
<h3 id="1-data-science-and-machine-learning">1. Data Science and Machine Learning</h3>
<p>Python’s rise in popularity can be attributed to its rich set of packages and tools for data science and machine learning. Researchers use data analysis packages like <a href="https://pandas.pydata.org/">pandas</a> to analyze <a href="https://www.sciencedirect.com/science/article/pii/S2352340920310726">Covid data</a>, practitioners can quickly apply machine learning with libraries like <a href="https://scikit-learn.org/">scikit-learn</a>, <a href="https://keras.io/">Keras</a>, and <a href="https://pytorch.org/">PyTorch</a> to detect fraud, classify images, and produce product recommendations.</p>
<h3 id="2-desktop-and-mobile-software-development">2. Desktop and Mobile Software Development</h3>
<p>Whether building simple software applications like calculators, or more <a href="https://codeinstitute.net/blog/7-popular-software-programs-written-in-python/">complex</a> ones like document editors and social networks. Python offers a plethora of tools to prototype, develop, and run desktop or mobile applications such as <a href="https://wxpython.org/">wxWidgets</a> and <a href="https://kivy.org/">Kivy</a>.</p>
<h3 id="3-web-development">3. Web Development</h3>
<p>Did you know that web applications like Uber, Netflix, and Spotify <a href="https://yourstory.com/mystory/top-twelve-python-applications-real-world-know">were built with Python</a>? There are a variety of web frameworks on Python that simplify website development. Packages such as <a href="https://flask.palletsprojects.com/">Flask</a>, <a href="https://www.djangoproject.com/">Django</a>, <a href="https://bottlepy.org/">Bottle</a>, and <a href="https://trypyramid.com/">Pyramid</a> all offer tools to quickly prototype and build scalable web apps.</p>
<h3 id="4-game-development">4. Game Development</h3>
<p>This may come as a surprise, but games such as The Sims 4 and Civilization IV were built with Python. Python libraries such as <a href="https://www.pygame.org/download.shtml">PyGame</a> and <a href="https://docs.python.org/3/library/turtle.html">Turtle</a> simplify prototyping and development of games.</p>
<h3 id="5-internet-of-things-iot-">5. Internet of Things (IoT)</h3>
<p>The rise of embedded systems and the internet of things (IoT) has catalyzed innovation and business process improvement across domains. Python offers a set of packages that make it easy to design, create, and deploy an IoT device on <a href="https://www.arduino.cc/">Arduino</a>, <a href="https://www.raspberrypi.org/">Raspberry Pi</a>, and other <a href="https://www.home-assistant.io/integrations/">IoT devices</a> such as Alexa and Google Home.</p>
<h2 id="6-steps-for-your-python-learning-journey">6 steps for your Python learning journey</h2>
<p><center><a href="https://datacamp.com"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606924547/6_steps_for_your_Python_learning_journey_-_no_title_kvijhj.png" /></a> <br></p>
<h3 id="1-spend-as-little-time-as-possible-on-basic-syntax">1. Spend as little time as possible on basic syntax</h3>
<p>When learning any programming language, it’s very easy to get bogged down by learning syntax. Introductory courses often make learning a chore by focusing on pure Python syntax rather than how Python can solve a specific problem. Don’t get me wrong: Learning Python syntax is essential for any beginner Python learner, but as a rule of thumb, you should learn just enough syntax to get started working on simple projects. When I began my Python journey for data science, I learned just enough syntax to be able to use <a href="https://pandas.pydata.org/">pandas</a> for simple data analysis tasks. Here’s a list of syntax resources:</p>
<ul>
<li><a href="https://realpython.com/products/python-basics-book/">A Practical Introduction to Python 3</a></li>
<li><a href="https://docs.python.org/3/tutorial/">Python’s Main tutorial page</a></li>
<li><a href="https://www.coursera.org/learn/python-crash-course">Coursera&#39;s Crash Course in Python taught by Google</a></li>
<li><a href="http://www.datacamp.com/courses/intro-to-python-for-data-science">DataCamp’s Introduction to Python course</a></li>
</ul>

<h3 id="2-set-up-your-python-environment">2. Set up your Python environment</h3>
<p>Before working on projects, you need to have a Python environment set up on your machine. Being able to work on projects on your own machine is the best way to apply your new skills. To set up your environment for data science, I recommend installing an <a href="https://www.anaconda.com/products/individual">Anaconda</a> distribution, which conveniently installs all relevant data science packages, and an integrated development environment for editing code in Python called <a href="https://www.spyder-ide.org/">Spyder</a>.</p>
<p>For a primer on setting up your Python environment with an Anaconda, watch my <a href="https://www.datacamp.com/resources/webinars/setting-up-your-python-environment">hands-on training</a> for a step-by-step tutorial. You can also follow beginner guides for installing Anaconda on <a href="https://www.datacamp.com/community/tutorials/installing-anaconda-windows">Windows</a> and <a href="https://www.datacamp.com/community/tutorials/installing-anaconda-mac-os-x">Mac OS X</a>.</p>
<p><center><a href="https://www.spyder-ide.org/"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606331451/spyderIDE_sakabc.png" /></a></p>
<h5 id="an-image-of-the-spyder-integrated-development-environment-https-www-spyder-ide-org-">An image of the <a href="https://www.spyder-ide.org/">Spyder Integrated Development Environment</a></h5>
<h3 id="3-learn-by-doing">3. Learn by doing</h3>
<p>At DataCamp, we believe that learning by doing is the best way to learn and apply programming skills. No matter your focus area, working on projects will help you get out of your comfort zone, learn how to debug and read documentation, and showcase your skills to potential employers.</p>
<h4 id="data-science-and-machine-learning">Data Science and Machine Learning</h4>
<p>Data Science and Machine Learning is one of the <a href="https://business.linkedin.com/content/dam/me/business/en-us/talent-solutions/emerging-jobs-report/Emerging_Jobs_Report_U.S._FINAL.pdf">fastest-growing</a> focus areas today. There are many solutions for online data skills training:</p>
<ul>
<li><a href="http://datacamp.com/">DataCamp</a> offers over 130 interactive Python courses and 39 projects across a wide range of data science and machine learning topics. You’ll build your skills from the ground up by coding directly in the browser, and apply your skills on a variety of datasets from Finance, Marketing, Sports, History and more.</li>
<li>The <a href="https://jakevdp.github.io/PythonDataScienceHandbook/">Python Data Science Handbook</a> provides an excellent overview of the main data science packages and techniques in Python. Conveniently, all of the content is available on <a href="https://github.com/jakevdp/PythonDataScienceHandbook">GitHub</a> in the form of Jupyter notebooks.</li>
<li>The <a href="https://www.coursera.org/professional-certificates/ibm-data-science">IBM Data Science Professional Certificate</a> specialization on Coursera contains nine courses covering a range of techniques in data analysis, data visualization, machine learning, and more. The <a href="https://www.coursera.org/learn/applied-data-science-capstone">final course</a> in the specialization is a capstone project combining these techniques on Foursquare data.</li>
<li>The <a href="https://www.codecademy.com/learn/paths/data-science">Codeacademy Data Scientist Path</a> offers a range of courses on data science and machine learning with Python, culminating in a final portfolio project.</li>
</ul>

<h4 id="desktop-and-mobile-software-development">Desktop and Mobile Software Development</h4>
<p>Python offers a range of tools to prototype and develop desktop and mobile applications. Here’s a list of learning resources to get you started:</p>
<ul>
<li>The <a href="http://zetcode.com/wxpython/">ZetCode wxWidgets tutorial</a> offers a step-by-step introduction to building graphical user interfaces with <a href="https://wxpython.org/">wxWidgets</a> and shows learners how to build a simple Tetris game.</li>
<li><a href="https://kivy.org/">Kivy</a> offers a range of tutorials on how to build simple mobile applications with Kivy on Python.</li>
</ul>

<h4 id="web-development">Web Development</h4>
<p>The two main packages for Python web development are Flask and Django. While there is <a href="https://medium.com/@SteelKiwiDev/flask-vs-django-how-to-understand-whether-you-need-a-hammer-or-a-toolbox-39b8b3a2e4a5">discussion</a> on the pros and cons of each, here’s a list of learning resources for both packages:</p>
<ul>
<li>The <a href="https://flask.palletsprojects.com/en/1.1.x/tutorial/">Flask tutorial page</a> explains how to build a simple blog website. If you’re a visual learner, <a href="https://www.youtube.com/watch?v=MwZwr5Tvyxo">this YouTube video</a> is an excellent introduction to building websites with Flask.</li>
<li>Learn to build web apps with the <a href="https://docs.djangoproject.com/en/3.1/intro/tutorial01/">Django tutorial page</a>. <a href="https://realpython.com/get-started-with-django-1/">Real Python</a> also explains this nicely, and includes a list of sample projects for a web developer portfolio.</li>
</ul>

<h4 id="game-development">Game Development</h4>
<p>While modern game development is dominated by game engines like Unreal Engine and Unity where most developers use JavaScript, C#, C++ or Java, the following python packages are a great way to get into game development:</p>
<ul>
<li>The <a href="http://www.pygame.org/wiki/tutorials">PyGame tutorial page</a> shows you how to build a range of different games with PyGame. Visual learners can follow <a href="https://www.youtube.com/watch?v=Q-__8Xw9KTM">this YouTube video</a> to build a space invaders game with Python.</li>
<li><a href="https://realpython.com/beginners-guide-python-turtle/">Real Python</a> offers an excellent overview of the <a href="https://docs.python.org/3/library/turtle.html">Turtle</a> package for game development in Python, ending with a turtle race game as a final project.</li>
</ul>

<h4 id="internet-of-things-iot-">Internet of Things (IoT)</h4>
<p>IoT devices are everywhere around us, collecting data about our environment and are only expected to rise in numbers, here’s a list of resources covering how to program IoT devices with Python:</p>
<ul>
<li>Coursera&#39;s <a href="https://www.coursera.org/specializations/iot">Introduction to Programming the Internet of Things (IOT) Specialization</a> offers a series of courses for beginners on how to interface with the Arduino and Raspberry Pi platforms with Python. The specialization ends with a <a href="https://www.coursera.org/learn/internet-of-things-project">capstone project</a> where learners will program an IoT system.</li>
<li>LinkedIn Learning’s <a href="https://www.linkedin.com/learning/internet-of-things-with-python-and-raspberry-pi/the-course-overview">Internet of Things with Python and Raspberry Pi</a> shows you how to program Raspberry Pi with Python by using different projects as case studies.</li>
</ul>

<h3 id="4-learn-how-to-debug-code-and-ask-for-help">4. Learn how to debug code and ask for help</h3>
<p>One of the most overlooked skills any aspiring programmer needs to hone is debugging. Every Python practitioner, regardless of skill level, faces errors and problems with their code. When you encounter with errors in your code, make sure to use these resources at your disposal:</p>
<ul>
<li><a href="http://google.com/">Google</a> is almost always the first place I go when faced with errors with my code. Almost always, the answer to my Python query will be in the first 10 search results.</li>
<li><a href="https://stackoverflow.com/">Stack Overflow</a> is the largest community of developers where practitioners can seek help with debugging code by asking and answering programming questions. Most of the time, your Python question has already been <a href="https://stackoverflow.com/questions/tagged/python">asked and answered</a>. If you want to post a question yourself, make sure to follow their <a href="https://stackoverflow.com/help/how-to-ask">best practices when phrasing questions</a>.</li>
<li>Documentation will always be your friend. Whether working with built-in <a href="https://docs.python.org/3/">Python</a> functions or working with a downloaded package like <a href="https://pandas.pydata.org/docs/">pandas</a>, you can always find documentation detailing functionality, possible errors, and tutorials on the website of the package you’re working with.</li>
</ul>

<h3 id="5-expand-your-learning-resources">5. Expand your learning resources</h3>
<p>Now that you’ve set up your Python environment, worked on guided projects, and learned how to debug, it’s time to expand your knowledge and continuously sharpen your skills. Here are a list of resources that will help you do that:</p>
<ul>
<li>Blogs can help you keep up with the latest trends and techniques in Python. One of my favorites is <a href="https://realpython.com/">Real Python</a>, as it hosts primers and tutorials on packages I wouldn’t otherwise use. They also have a bit-sized newsletter with Python tips and tricks.</li>
<li>Reading other people’s code is immensely useful when sharpening your Python skills. For data science and machine learning, <a href="https://www.kaggle.com/">Kaggle</a> is an excellent resource to see how experienced data scientists would solve a problem. You can also look at the source code of your favorite open-source packages on GitHub.</li>
<li>Attend <a href="https://python.geekle.us/">conferences</a>, <a href="https://www.meetup.com/topics/python/">meetups</a>, and other community gatherings with more experienced practitioners. This will give you the opportunity to network and learn from more experienced Python programmers.</li>
</ul>

<h3 id="6-build-a-portfolio-of-projects">6. Build a portfolio of projects</h3>
<p>Now you&#39;re ready to start tackling complex open-ended projects and showcase them in a portfolio for future employers. When working on open-ended projects, it&#39;s best to start simple  and work on Python projects directly related to your work. If you&#39;re looking for inspiration, <a href="https://www.edureka.co/blog/python-projects/">this Edureka blog post</a> contains a list of open-ended projects in game development, web development, data science and machine learning.</p>
<p>You can start building your portfolio by hosting your code on <a href="http://www.github.com">Github</a> and writing blog posts about your projects. This gives you the ability to showcase your technical writing ability and teach beginners how you&#39;ve developed your projects. You can write blog posts by developing your own website or by using a blog hosting service like <a href="http://www.medium.com">Medium</a>.</p>
<p>I hope you feel inspired by this blog post, and that these resources are useful to you. Mastering any programming language is a lifelong learning journey—one that I am still on myself. At DataCamp, Python is part of our core curriculum and we’re constantly improving and adding to our content. Lastly, I invite you to join thousands of learners in the <a href="https://support.datacamp.com/hc/en-us/articles/360001915993-The-DataCamp-Slack-Community-An-Overview">DataCamp Slack Community</a> to get support on your Python journey!</p>
<p><em>Adel Nehme is a Data Science Evangelist at DataCamp. He was previously a Content Developer where he collaborated with instructors on a range of Python courses, and created our course <a href="http://datacamp.com/courses/cleaning-data-in-python">Cleaning Data in Python</a>.</em></p>


https://www.datacamp.com/community/blog/how-to-learn-python

Here’s all you need to know to get started with Python.

How to Learn Python

<p><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605885235/Fast_500_ca58zo.png" /></a></p>
<p><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_1_cruuf7.png" /></a><br></p>
<p>DataCamp is proud to announce that for the second year in a row, we’ve been recognized on Deloitte’s Technology Fast 500™, a ranking of the 500 fastest-growing technology, media, telecommunications, life science, fintech, and energy tech companies in North America. DataCamp ranked 94th this year, with a growth rate of 1,232% over the three-year period from 2016 to 2019, and is the second-fastest-growing edtech company to have earned this distinction.</p>
<p>DataCamp CEO and co-founder Martijn Theuwissen credits our growth to the evolution of our platform and curriculum, and the crucial role of online learning in digital transformation initiatives.</p>
<p>“We are delighted to be recognized for the second year in a row on Deloitte’s Technology Fast 500 ranking of the fastest-growing companies in North America,” said Theuwissen. “With nearly seven million learners and more than 1,600 business customers, DataCamp enables companies to build data fluency at scale. Data skills aren’t just for technical roles anymore—the most successful companies know that today, everyone needs data skills to do their jobs effectively.</p>
<p>“In addition to providing a best-in-class learning solution to build these data skills, DataCamp also allows companies to track their teams’ skill progress, measure program adoption and engagement, compare their progress with industry peers, and identify organizational strengths and skill gaps. We understand that data is at the core of every business today, which is why we are committed to delivering companies and individual learners the world’s best platform to become data fluent.”</p>
<h2 id="our-platform-and-curriculum">Our Platform and Curriculum</h2>
<h3 id="datacamp-for-business">DataCamp for Business</h3>
<p>We help organizations create custom learning journeys based on their teams’ skill sets and desired learning outcomes. Data-driven organizations aspire to be data-fluent at every level, which means everyone is equipped to answer business questions with data. The success of digital transformation initiatives pivots on having the appropriate data skills across the organization.</p>
<p><a href="https://www.datacamp.com/groups/business">DataCamp for Business</a> enables companies to build these data skills in every role and at every skill level. It’s easy to implement and manage for teams of any size, with an admin dashboard that allows custom learning paths based on roles and departments, <a href="https://www.datacamp.com/groups/business/reporting">Advanced Enterprise Reporting</a> to measure the impact of online learning, <a href="https://www.datacamp.com/groups/business/customer-success">expert guidance from a dedicated Customer Success Manager</a>, and <a href="https://www.datacamp.com/groups/business/integration">seamless SSO and LMS integrations</a>. And we’re laying the groundwork for organizations to visualize their team’s strengths and skill gaps at scale with an automated skill matrix.</p>
<h3 id="the-most-effective-platform-for-data-skill-training">The most effective platform for data skill training</h3>
<p>DataCamp’s proven learning methodology helps companies and individuals systematically build and apply the data skills they need to make better business decisions.</p>
<p><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606140563/ALPA_1_dfcgmq.png" /></a></p>
<h4 id="assess">Assess</h4>
<p>Effective learning starts with assessing learners’ skills to identify strengths and skill gaps. Organizations can leverage <a href="http://datacamp.com/signal">DataCamp Signal™</a> to provide personalized learning recommendations, save time by identifying exactly what each learner needs to improve, and focus on the skills needed to accomplish their business objectives.</p>
<p>We now have 14 total assessments on topics in Python, R, SQL, and understanding and interpreting data—with many more on the way. <a href="https://www.datacamp.com/resources/whitepapers/datacamp-signal">Learn more</a> about the methodology behind what makes Signal the gold standard in applied data skill assessment.</p>
<h4 id="learn">Learn</h4>
<p>We’ve grown our curriculum to cover a wide breadth of technologies and topics, including new content for people in non-technical roles who need to be able to analyze and interpret data and work more effectively with data professionals. We’ve created content in key topics like <a href="https://www.datacamp.com/tracks/machine-learning-for-everyone">Machine Learning for Everyone</a> and <a href="https://www.datacamp.com/courses/data-visualization-for-everyone">Data Visualization for Everyone</a>, and skill tracks like <a href="https://www.datacamp.com/tracks/foundational-data-skills-for-business-leaders">Data Skills for Business</a> and <a href="https://www.datacamp.com/tracks/data-literacy-fundamentals">Data Literacy Fundamentals</a>. We’ve also increased content centered on business intelligence tools like <a href="https://www.datacamp.com/courses/analyzing-data-in-tableau">Tableau</a>, <a href="https://www.datacamp.com/courses/introduction-to-power-bi">Power BI</a>, and <a href="https://www.datacamp.com/courses/data-analysis-in-excel">Excel</a>, and expanded our focus on <a href="https://www.datacamp.com/tracks/data-engineer-with-python">data engineering</a> to help you collect, clean, and catalog your business data. We’ve also made learning more accessible with subtitles in nine new languages!</p>
<p>Our library of interactive courses has grown from just one course on January 1, 2015, to over 340 courses today, and our average course completion rate of 60% is far higher than the industry average of just 15%.</p>
<p><center><a href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1606140564/Number_of_DataCamp_courses_wn6ree.png" /></a></p>
<h4 id="practice-and-apply">Practice and Apply</h4>
<p>The DataCamp <a href="https://www.datacamp.com/mobile/">mobile app</a> is now fully optimized for the best possible coding experience on mobile with features like autocomplete, making it easier than ever to learn how to code on a mobile device. The mobile learning experience is also now structured around lessons as opposed to full courses, making it easier to make progress with shorter units.</p>
<p><a href="http://datacamp.com/projects">Unguided projects</a> provide a new way to apply professional data science skills and build the confidence learners need to code on their own in the real world. Unguided projects are designed for intermediate and advanced learners who want to solve open-ended data science problems without step-by-step tasks. And if they get stuck, learners can follow a live code-along video to see how our expert instructors find one of many possible solutions.</p>
<h2 id="our-customers">Our Customers</h2>
<p>We’re on a mission to make data fluency accessible to millions of people and businesses around the world. We’ve grown our learners to nearly 7 million, and have more than 1,600 business customers, including Google, Intel, HSBC, eBay, PayPal, Uber, and more. And through DataCamp for the Classroom, <a href="https://www.datacamp.com/groups/education">we’ve provided free access to DataCamp</a> to more than 350,000 students from 180 countries.</p>
<h2 id="our-partnerships">Our Partnerships</h2>
<p>DataCamp is committed to supporting data education for all. As part of our mission to democratize data skills for everyone, we currently have over <a href="https://support.datacamp.com/hc/en-us/articles/360051596574">30 nonprofit partners</a>—like <a href="https://www.datacamp.com/community/blog/datau">DataU</a>, <a href="https://www.datacamp.com/resources/case-studies/blossom-academy">Blossom Academy</a>, and <a href="https://www.datacamp.com/resources/case-studies/duke">Duke University&#39;s Fuqua Volunteer Corps</a>—that we'll provide with free access to DataCamp for more than 25,000 learners around the world. Our focus on supporting these communities has only become more critical during a global pandemic, when now more than ever, learners need to build data fluency for a more secure future.</p>
<p>“For a lot of students, it’s easy in a pandemic to just feel lost,” said Jeremy Petranka, Assistant Dean of Duke Fuqua’s Master of Quantitative Management (MQM) Program. “It was important for us that they feel supported by Fuqua and by DataCamp. Being able to say early on that DataCamp was behind them made them feel really valued.”</p>
<h4 id="to-learn-more-about-datacamp-being-named-a-technology-fast-500-award-winner-read-the-press-release-">To learn more about DataCamp being named a Technology Fast 500™ award winner, <a href="https://www.prnewswire.com/news-releases/datacamp-ranked-94th-fastest-growing-company-in-north-america-on-deloittes-2020-technology-fast-500-301179267.html">read the press release</a>.</h4>
<h2 id="about-deloitte-s-2020-technology-fast-500-">About Deloitte’s 2020 Technology Fast 500™</h2>
<p>Now in its 26th year, Deloitte’s Technology Fast 500™ is a ranking of the 500 fastest-growing technology, media, telecommunications, life sciences, and energy tech companies in North America. Overall, 2020 Technology Fast 500™ companies achieved revenue growth ranging from 175% to 106,508% from 2016 to 2019, with a median growth of 450%.</p>
<p>In order to be eligible for Technology Fast 500 recognition, companies must own proprietary intellectual property or technology that is sold to customers in products that contribute to a majority of the company&#39;s operating revenues. Companies must have base-year operating revenues of at least \$50,000 USD, and current-year operating revenues of at least \$5 million USD. Additionally, companies must be in business for a minimum of four years and be headquartered within North America.</p>
<p><center><a href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p>


https://www.datacamp.com/community/blog/fast-500-2020

DataCamp grew 1,232% from 2016 through 2019.

DataCamp Is in the Top 100 Fastest Growing Companies on Deloitte’s Technology Fast 500™ 

<p><center><a href="https://www.datacamp.com/resources/whitepapers/digital-transformation-in-finance"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605539369/Download_white_paper_bnktg9.png" /></a><br></p>
<p>Over the past decade, big data and digital technologies have disrupted industries and consumer behavior alike. IDC and Statista estimate that the volume of data generated yearly rose from two zettabytes in 2010 to 59 zettabytes in 2020, marking a thirtyfold increase in data generated in the past 10 years alone (<a href="https://www.statista.com/statistics/871513/worldwide-data-created/">Statista</a>). This data deluge is only expected to grow, with projections predicting 149 zettabytes produced yearly by 2024.</p>
<p>While various industries are vying to take advantage of the data deluge with business intelligence, data science, and machine learning, the financial services industry is best equipped to benefit from big data. Data is at the heart of the financial services industry—across retail banking, investment banking, and insurance. Financial services organizations produce and store data on their customer transactions, detailed customer profiles through compliance processes, insurance claims, stock market exchanges, and more. The amount of data generated is astounding: The New York Stock Exchange alone produces one terabyte of trade data daily (<a href="https://www.investopedia.com/articles/active-trading/040915/how-big-data-has-changed-finance.asp">Investopedia</a>).</p>
<p>We’ve already seen fintech startups take advantage of shifting consumer behavior and the financial industry’s data deluge. Digital banks such as N26, Revolut, and Monzo abandoned the brick-and-mortar model and opted for a purely digital banking experience, relying on data to improve user experience and automate workflows (<a href="https://blog.revolut.com/how-we-use-machine-learning-to-protect-you-from-fraud/">Revolut</a>). Klarna, Europe’s most giant fintech unicorn, provides interest-free installment options with automated approval or rejection using machine learning (<a href="https://www.cnbc.com/2020/09/14/klarna-now-europes-biggest-fintech-unicorn-at-over-10-billion-value.html#:~:text=CNBC%20Disruptor%2050-,Klarna%20becomes%20Europe&#39;s%20biggest%20fintech%20unicorn%20at,billion%20valuation%20after%20mega%2Dround&amp;text=Klarna%20nearly%20doubled%20its%20prior,%245.5%20billion%20to%20%2410.6%20billion.">CNBC</a>). The data deluge has not only opened up space for disruptive, innovative services—it’s opened the door for data-enabled digital transformation across the industry.</p>
<p>Disruptive digital-first startups across all industries have prompted many incumbents to invest heavily in digital transformation. The financial services industry is no exception. An Accenture and Oxford study in 2018 found that 87% of retail banking executives have developed a long-term plan for technology investment and digital transformation (<a href="https://www.accenture.com/_acnmedia/pdf-92/accenture-retail-banks-well-prepared-for-digital-innovation-infographic.pdf">Accenture</a>). This is especially true in the COVID-19 economy, which has moved consumers purchasing online and accelerated digital transformation programs across all industries.</p>
<p>This acceleration is exceptionally pressing in the financial services industry. A <a href="https://eiuperspectives.economist.com/financial-services/forging-new-frontiers-advanced-technologies-will-revolutionise-banking">recent study from the Economist Intelligence Unit</a> cites that 45% of banking executives believe building a “true digital ecosystem” is the best strategic response to the pandemic. In the same survey, 66% of respondents believe that new technologies such as machine learning and artificial intelligence will bring the most significant impact on the banking industry by 2025.</p>
<p>Taking an example from the ground, the urgency of using contact-less financial tools ushered an 84% increase in Citibank’s daily mobile check deposits, and a tenfold increase in activity on Apple pay (<a href="https://www.forbes.com/sites/antoinegara/2020/06/08/the-worlds-best-banks-the-future-of-banking-is-digital-after-coronavirus/">Forbes</a>). This has prompted Jane Fraser, president of Citigroup and CEO of its consumer bank, to declare, “Banking has changed irrevocably as a result of the pandemic. The pivot to digital has been supercharged. [...] We believe we have the model of the future—a light branch footprint, seamless digital capabilities, and a network of partners that expand our reach to hundreds of millions of customers.”</p>
<p><center><a href="https://www.datacamp.com/resources/whitepapers/digital-transformation-in-finance"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605545374/FINANCE_use_cases_br0ol6.png" /></a></p>
<p>The success of such digital transformation programs pivots on the seamless integration of digital technologies with data-driven insights and high-impact data science use-cases. What are these high-impact use cases and what are the challenges standing in the way? In our white paper, <a href="https://www.datacamp.com/resources/whitepapers/digital-transformation-in-finance">Digital Transformation in Finance: Upskilling for a data-driven age</a>, we dissect 13 high-impact use cases spread across domain and sector and the challenges large financial institutions face to becoming data-driven.</p>
<p><center><a href="https://www.datacamp.com/resources/whitepapers/digital-transformation-in-finance"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605539369/Download_white_paper_bnktg9.png" /></a></p>


https://www.datacamp.com/community/blog/digital-transformation-in-finance

Financial institutions can seamlessly integrate digital technologies with data-driven insights.

Find out how financial institutions can seamlessly integrate digital technologies with data-driven insights.

13 Use Cases for Data-Driven Digital Transformation in Finance

<center><a href="https://www.datacamp.com/resources/webinars/thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604688720/Watch_recording_qn82wo.png" /></a></center><br>

<p>As organizations produce more data and digitize products and processes, a data-driven workforce has never been more critical. This is why learning and development has become central to business strategies—especially initiatives focused on building organization-wide data science capabilities.</p>
<p>On November 4, DataCamp’s Data Science Evangelist, Adel Nehme, was joined by Sheil Naik, Global Data Technical Trainer at Bloomberg, to discuss how Sheil’s team leverages blended learning to teach data analysis with Python, and how Bloomberg measures behavioral change following their upskilling initiatives.</p>
<h2 id="the-value-of-becoming-data-driven">The value of becoming data-driven</h2>
<p>Never before has it been more valuable to be data-driven. Anaconda’s CEO Peter Wang describes data science as an evidence-based methodology for solving business problems, where data scientists “harness mathematical and computational tools to reason about the business world.” This methodology has enabled a plethora of use cases across industries, from forecasting churn in marketing to automatic fraud detection for financial institutions.</p>
<p>Data science is at the heart of Bloomberg’s data-driven culture. Their data science team is pushing the boundaries of what’s possible in many ways including creating <a href="https://www.techatbloomberg.com/blog/bloombergs-global-data-cto-data-science-teams-publish-best-practices-for-data-annotation-projects/">best practices</a> for natural language processing projects, <a href="https://www.techatbloomberg.com/blog/bqplot-is-now-a-community-open-source-project/">democratizing</a> data tools, and providing <a href="https://www.bloomberg.com/company/press/bloomberg-magic-machine-learning/">intelligent solutions</a> across its products. Having a data-driven culture also means enabling everyone with the necessary skills to make data-driven decisions, improve processes with data, and produce data-driven news stories.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605042163/image_1_ylkzmq.png" /></center>

<h2 id="building-a-data-driven-workforce-with-blended-learning">Building a data-driven workforce with blended learning</h2>
<p>The Global Data Division at Bloomberg is responsible for maintaining the timeliness and quality of all financial datasets found on the Bloomberg Terminal. As a data technical trainer, Sheil Naik works with business leaders across the Division to identify the skills needed to be successful when working with data at Bloomberg and to design, deliver, and evaluate training programs aimed at building and improving these skills. These skills include using version control tools like Git and GitHub, data analysis with SQL, data analysis with Python, and more.</p>
<p>Bloomberg’s Data Analysis with Python program is a quarterly, blended-learning curriculum incorporating a one-hour introduction explaining how Python is used at Bloomberg, 12 to 20 hours of DataCamp coursework, three live 1.5-hour sessions led by in-house technical experts, and a final project using Bloomberg data.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605042163/image_2_p3h2xf.png" /></center><br>

<p>Bloomberg carefully curates the DataCamp courses to balance learning objectives and time commitment and contain chapters from courses like <a href="https://www.datacamp.com/courses/intro-to-python-for-data-science">Introduction to Python</a>, <a href="https://learn.datacamp.com/courses/intermediate-python">Intermediate Python</a>, and more. Acting as foundational material, these courses have enabled more than 450 learners—many of whom never coded in their lives before—to learn and apply the concepts needed to complete the three live classroom sessions, pass the final project, and finish the training program.</p>
<blockquote>
<p>Learners were able to go from never writing a line of code in their entire life to completing a data-driven news analysis as part of the final project of the program—Sheil Naik, Global Data Technical Trainer at Bloomberg</p>
</blockquote>
<p>Combining self-guided learning with live classroom training, this blended learning model allows for consistency and flexibility of learning across geographies, schedules, and business units at Bloomberg. The consistency of the curriculum and experience provided at the foundational level allows for Bloomberg trainers to scale classroom sessions and include global learners. Moreover, using a learning provider like DataCamp at the foundational level offers insights and performance data used to gauge training effectiveness.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605042603/image_3_qpsdo6.png" /></center>

<h2 id="measuring-behavioral-change">Measuring behavioral change</h2>
<p>A methodology for evaluating direct return on investment for training programs is the Kirkpatrick Model of Evaluation. The Kirkpatrick Model proposes four different evaluation levels: the initial reaction following a training program, learning evaluation, behavioral change, and the business impact of gained skills.</p>
<p>Bloomberg’s implementation of the third layer of the Kirkpatrick model measures the number of producer activities (saves, edits, imports, renames, sends, etc…) on their proprietary Jupyter Notebook environment.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605042163/image_4_vz0iz0.png" /></center><br>

<p>By leveraging these data points, Sheil was able to apply the techniques taught in Data Analysis with Python to uncover a <strong>561%</strong> increase in average producer activities for one of the cohorts graduating from the program. According to Sheil, the ultimate goal of data upskilling at Bloomberg is to combine technology with employees’ subject matter expertise to produce insightful analyses.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1605042163/image_5_pb51sc.png" /></center><br>

<p>If you want to learn more about blended learning, how to operationalize it in your own organization, and the key takeaways Sheil recommends every learning and development professional to follow, make sure to <a href="https://www.datacamp.com/resources/webinars/thrive-in-a-data-driven-age">watch the full webinar recording</a>.</p>
<br><center><a href="https://www.datacamp.com/resources/webinars/thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604688720/Watch_recording_qn82wo.png" /></a></center>


https://www.datacamp.com/community/blog/data-driven-learning

Sheil Naik, Global Data Technical Trainer at Bloomberg, explains how organizations can use a blended learning program to scale data upskilling and measure behavioral change of training initiatives.

Scaling Blended Learning at Bloomberg 

<h1 id="our-mobile-app-has-gotten-a-facelift-">Our mobile app has gotten a facelift!</h1>
<p>DataCamp Mobile has a new and improved user experience. Instead of working through full length courses designed for a desktop computer, mobile learning is now structured around lessons—so you can see what you’ll learn at a glance. And even if you just have five minutes to spare, you’ll be able to measure how much progress you’ve made. This is a big step for learner motivation since you’ll be able to make progress with shorter units.</p>
<p>We’ve launched a bunch of new courses in SQL, R, and Python, including the complete tracks for Intro to SQL and Intro to R.</p>
<iframe width="560" height="315" src="https://www.youtube.com/embed/dir9FHN80ag" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

<h1 id="run-real-code-on-your-mobile-device">Run real code on your mobile device</h1>
<p>Our mobile courses give you the full coding experience—they’re fully equivalent to our official courses. When you open the mobile app, you’ll access the full editor and we’ll spin up a server for you to run real code and see real results. You’re writing code on your phone for a fully interactive experience—this isn’t just multiple choice. With the assistance of a mobile-optimized code editor (including autocomplete), mobile learning has never been more effective.</p>
<blockquote>
<p>Finally, an app that actually makes coding fun. —Navistonv</p>
</blockquote>
<h1 id="seamless-learning-from-anywhere">Seamless learning from anywhere</h1>
<p>Modern learning requires flexibility and convenience, which our standalone mobile app provides in spades. With DataCamp for Mobile, we’re providing a holistic learning solution. All progress on DataCamp courses syncs seamlessly to your account. You can start a course on your desktop and pick up where you left off on our mobile app. It’s a truly mobile-friendly way to learn.</p>
<blockquote>
<p>Now this is how things should be done! —pyggh</p>
</blockquote>
<p><center><a href="https://datacamp.app.link/MKQyDIGRabb"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604678489/DataCamp_for_Mobile_Blog_xojyv8.png" /></a><br></p>
<h1 id="a-fun-way-to-keep-your-skills-sharp">A fun way to keep your skills sharp</h1>
<p>Practice makes perfect, and you’ll want to make a habit of exercising your new skills. That’s why we’ve gamified learning so it doesn’t feel like work. You’ll accumulate experience points and get real-time feedback. Challenge your friends or colleagues to see how many days you can extend your learning streak.</p>
<blockquote>
<p>Excellent on-the-go practice. —Gray0417</p>
</blockquote>
<h1 id="get-started">Get started</h1>
<p><a href="https://datacamp.app.link/MKQyDIGRabb">Get started on mobile</a> with our Data Scientist in Python, Data Scientist in R, and SQL Fundamentals tracks, with even more courses coming soon!</p>
<p><center><a href="https://datacamp.app.link/MKQyDIGRabb"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604677518/Mobile_Banner_jn5jjc.png" /></a><br></p>


https://www.datacamp.com/community/blog/datacamp-for-mobile

To help you learn more effectively, DataCamp has combined the convenience of mobile learning with a robust coding infrastructure. 

DataCamp for Mobile: Mobile Coding at Its Best

<p>In the past several months, the data science and AI worlds have been abuzz with the arrival of GPT-3, OpenAI’s newest language model. For many, the model represents a significant leap in the ability of one algorithm to reason with human language across a variety of tasks.</p>
<p>Developers testing GPT-3 have provided many intriguing use cases. Examples of automatic code generation based on plain English prompt, answering medical questions, and legal language translation have ignited the imaginations of many data scientists thinking about the next generation of AI-powered software.</p>
<p><center><a href="https://twitter.com/sharifshameem/status/1284807152603820032?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604590183/Screen_Shot_2020-11-05_at_10.28.53_AM_eejh4q.png" /></a></center><br></p>
<p><center><a href="https://twitter.com/QasimMunye/status/1278750809094750211?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604590335/Screen_Shot_2020-11-05_at_10.32.11_AM_xqgdjo.png" /></a></center><br></p>
<p><center><a href="https://twitter.com/f_j_j_/status/1283349995144359937?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604590534/Screen_Shot_2020-11-05_at_10.35.27_AM_kgfk88.png" /></a></center><br></p>
<p>While much of the value of machine learning at an organizational level is in low-hanging fruit like predicting churn, simple sales forecasts, and customer segmentation—it’s useful to consider what the <a href="https://openai.com/blog/openai-api/">commercialization of GPT-3</a> means for the future. It has the potential to transform how we think about and operationalize artificial intelligence.</p>
<h2 id="defining-artificial-intelligence-machine-learning-and-deep-learning">Defining artificial intelligence, machine learning, and deep learning</h2>
<p>The business world and media are overloaded with buzz terms like artificial intelligence (AI), machine learning (ML), and deep learning (DL). Let’s quickly define these terms before delving into how GPT-3 works.</p>
<p>Andrew Ng, co-founder of Google Brain and former Chief Scientist at Baidu, describes artificial intelligence as a “huge set of tools for making computers behave intelligently.” This encompasses explicitly programmed software like calculators, as well as ML applications like recommendation systems and self-driving cars.</p>
<p>Machine learning is the “field of study that gives computers the ability to learn without being explicitly programmed,” according to Arthur Samuel, pioneer in artificial intelligence and computer gaming. There are generally two types of machine learning algorithms. The first is supervised learning, where algorithms learn the patterns between existing data (inputs) and labels (outputs) and predict the output on unseen data, like whether a new customer will churn based on historical churn data. The second is unsupervised learning, where algorithms discover general patterns in the data and cluster different data points that are similar to each other, as in segmenting customers based on common patterns of behavior.</p>
<p>Deep learning is a form of machine learning based on multi-layered artificial neural networks, which are loosely inspired by biological neural networks in the brain. They can be both supervised and unsupervised and are largely responsible for the last decade’s high-profile ML use cases, like image recognition and sentiment analysis. Deep learning models vary in architecture, ranging from simple to complex based on the number of layers and nodes in the neural network architecture. The more complex a model is, the more parameters it has. If you want to learn more about how deep learning models are built, check out DataCamp’s <a href="https://www.datacamp.com/tracks/deep-learning-in-python">deep learning skill track</a>.</p>
<p>For a more in-depth exploration of these topics, read our e-book, <a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders">The Definitive Guide to Machine Learning for Business Leaders</a>.</p>
<h2 id="how-gpt-3-works">How GPT-3 works</h2>
<p>So where does GPT-3 intersect with artificial intelligence, machine learning, and deep learning? The acronym GPT refers to “generative pre-trained transformer”—an unsupervised deep learning algorithm that is usually pre-trained on a large amount of unlabeled text. It’s fine-tuned and trained on a large task-specific labeled dataset (e.g., translation of English to French), and is then tasked with inferring the most likely set of outputs (French translation) given a specific set of inputs (English words). You can think of this as a highly sophisticated form of autocomplete for a variety of different language tasks.</p>
<p>GPT-3 is the third iteration of this model, and while it does not innovate on the architecture of its predecessors, it’s pre-trained on extremely large datasets comprising a large portion of the internet, including the <a href="https://commoncrawl.org/the-data/">Common Crawl</a> dataset, and includes many more layers in its network architecture. This makes GPT-3 the most complex language model ever conceived, with 175 billion parameters in its network architecture. This is ten times more parameters than the most complex model prior to GPT-3’s release, <a href="https://www.microsoft.com/en-us/research/blog/turing-nlg-a-17-billion-parameter-language-model-by-microsoft/">Turing-NLG by Microsoft</a>, and 117 times more complex than GPT-2.</p>
<p>Most importantly, GPT-3 benefits from few-shot learning, where the pre-trained model doesn’t have to be fine-tuned with large labeled training data for a specific language task. It’s instead just given a task description—translate English words to French—and a few examples of inputs mapped to outputs. Coupled with an easy to use plug-and-play interface, GPT-3 largely eliminates barriers to entry and allows non-experts to produce meaningful results on different language tasks.</p>
<p><center><a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1598020446/gpt3-1_gxwhsa.png" /></a></center><br></p>
<h2 id="why-gpt-3-is-important">Why GPT-3 is important</h2>
<p>With just a few examples and task descriptions, GPT-3 rivals fine-tuned language models that have been trained on task-specific training data for a variety of language tasks. GPT-3 also exhibits some form of success on tasks that require reasoning, like arithmetic, which are not necessarily language tasks. For example, GPT-3 exhibited 100% accuracy on two-digit addition and subtraction after it was fed a few examples of addition and subtraction. Less complex models with fewer parameters have not been able to break the 60% accuracy ceiling on these tasks. While GPT-3 falters on more complex forms of arithmetic, this implies that more complex models may have the ability to generalize outside of the domain they were trained on.</p>
<p><center><a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1598020446/gpt3-2_tiy26v.png" /></a></center><br></p>
<p>Interestingly, this suggests further gains can be achieved from purely increasing the dataset and model sizes. Currently, it doesn’t seem that the model’s aggregate performance across different tasks is plateauing at 175B parameters. Assuming the same increase in parameter scaling from GPT-2 to GPT-3, one can only wonder how model performance would scale if GPT-4 had 117 times more parameters than GPT-3.</p>
<p><center><a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1598020447/gpt3-3_krvb14.png" /></a></center><br></p>
<p>While it’s currently being calibrated in a private beta release, packaging GPT-3 in a plug-and-play API means it may be utilized at scale as soon as it’s out of private beta. As AI Researcher Shreya Shankar pointed out, an important challenge will be serving this API efficiently and easily for organizations to use.</p>
<p><center><a href="https://twitter.com/sh_reya/status/1284545976892403714?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604591445/Screen_Shot_2020-11-05_at_10.50.31_AM_jvfzw3.png" /></a></center><br></p>
<h2 id="what-this-means-for-the-future">What this means for the future</h2>
<p>New technologies often follow Gartner’s hype cycle—in fact, OpenAI CEO Sam Altman has already <a href="https://twitter.com/sama/status/1284922296348454913">sounded the hype alarm bells about GPT-3</a>.</p>
<p><center><a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1598020581/group_3x_ugeoch.png" /></a></center><br></p>
<p>However, the use cases coming out of GPT-3 developers shed light on the type of AI-powered applications we can expect in the medium to long term. Potential applications include tools that will help designers prototype easily, streamline data analysis, enable more robust research, automate content generation for content marketers, and more.</p>
<p><center><a href="https://twitter.com/dhvanilp/status/1286452207513038848?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604590877/Screen_Shot_2020-11-05_at_10.41.03_AM_kuc7gd.png" /></a></center><br></p>
<p><center><a href="https://twitter.com/FaraazNishtar/status/1285934622891667457?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604590984/Screen_Shot_2020-11-05_at_10.42.50_AM_yjls8y.png" /></a></center><br></p>
<p><center><a href="https://twitter.com/dhvanilp/status/1295661122079453184?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604591100/Screen_Shot_2020-11-05_at_10.44.43_AM_cn7ph9.png" /></a></center><br></p>
<p><center><a href="https://twitter.com/pavtalk/status/1285410751092416513?ref_src=twsrc%5Etfw"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604591198/Screen_Shot_2020-11-05_at_10.45.59_AM_flapcz.png" /></a></center><br></p>
<p>Additionally, packaging the model in an easy plug-and-play interface could change the dynamics of how AI is instrumentalized across the organization. For example, this could disincentivize organizations from developing their own in-house models, and allow less technical experts to build solutions using GPT-3.</p>
<p>Finally, when thinking about deploying AI systems at scale, you must be aware of the capacity to spread harm through bias. As many researchers noted while testing out GPT-3, it’s relatively easy to generate harmful outputs that reinforce stereotypes and biases based on neutral inputs.</p>
<p><center><a href="https://twitter.com/an_open_mind/status/1285940858290933767?s=20"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1604591295/Screen_Shot_2020-11-05_at_10.47.54_AM_wgqfxn.png" /></a></center><br></p>
<p>Like any machine learning algorithm <a href="https://www.datacamp.com/resources/ebooks/definitive-guide-to-machine-learning-for-business-leaders">being deployed at scale</a>, GPT-3 requires serious scrutiny and monitoring over potential harm.</p>
<p><center><a href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p>


https://www.datacamp.com/community/blog/gpt3

How GPT-3 expands the world of possibilities for language tasks—and why it will pave the way for designers to prototype more easily, streamline work for data analysts, enable more robust research, and automate content generation.

GPT-3 and the Next Generation of AI-Powered Services

<p><center><a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603892941/webinar_banner_hedvoq.png" /></a><br></p>
<p>As organizations produce more data and digitize products and processes, a data-driven workforce has never been more critical. This is why learning and development has become central to business strategies, especially initiatives focused on building data science capabilities across organizations.</p>
<p>Next week, DataCamp’s Data Science Evangelist, Adel Nehme, will host a <a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age">webinar</a> with Sheil Naik, Global Data Technical Trainer at Bloomberg, on how Bloomberg uses blended learning to teach data analysis with Python, and how these analysis techniques can be used to evaluate behavioral change and the success of upskilling initiatives.</p>
<h2 id="blended-learning-at-bloomberg">Blended learning at Bloomberg</h2>
<p>A data-driven workforce at Bloomberg uses data science to make data-driven business decisions, improves processes with data, and creates data-driven news stories. Sheil Naik leads a company-wide blended learning program focused on Data Analysis with Python which incorporates 12 to 20 hours of learning via DataCamp per quarter, three live 1.5-hour classroom sessions led by technical experts, and a final project using Bloomberg data as the final deliverable.</p>
<p><center><a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603894417/datadriven1_sjgl2a.png" /></a><br></p>
<p>The webinar will dive deep into the benefits of blended learning. The combination of DataCamp courses and virtual in-person classroom sessions allows for greater consistency and
flexibility across regional and scheduling constraints and increases the capacity of class sizes across the world.   </p>
<h2 id="evaluating-upskilling-success">Evaluating upskilling success</h2>
<p>A methodology for evaluating direct return on investment for training programs is the Kirkpatrick Model of Evaluation. In brief, the Kirkpatrick Model proposes four different evaluation levels: the initial reaction following a training program, learning evaluation, behavioral change, and the business impact of gained skills.</p>
<p><center><a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603898647/Train_Your_Workforce_to_Thrive_in_a_Data-Driven_Age_nwfelw.png" /></a><br></p>
<p>In the <a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age">webinar on November 4</a>, Sheil will deep dive into how Bloomberg operationalized the Kirkpatrick Model of Evaluation by measuring and evaluating post-training behavioral change for all learners in the program.</p>
<h2 id="reap-the-benefits-of-data-upskilling">Reap the benefits of data upskilling</h2>
<p>This only scratches the surface of Adel’s and Sheil’s discussion on creating a data-driven workforce. To find out more about how your organization can scale personalized learning, evaluate training success, and tie training to business objectives, <a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age">join our webinar on November 4th</a>.</p>
<p><center><a href="https://www.datacamp.com/webinars/train-your-workforce-to-thrive-in-a-data-driven-age"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603892941/webinar_banner_hedvoq.png" /></a><br></p>


https://www.datacamp.com/community/blog/blended-learning

Bloomberg’s Global Data Technical Trainer, Sheil Naik, breaks down how a blended learning environment can catalyze a data-driven workforce.

community

datacamp

Official Blog