AI is transforming cloud computing: How can devs stay ahead?
Every decade or so, we see a major paradigm shift in tech. When I first entered tech in the '00s, it was the cloud. Since then, cloud's emergence has pushed developers to learn and develop various new skills, from cloud computing fundamentals to distributed system design. In just a short span of time, the cloud completely transformed the tech industry as we know it — and demanded that developers follow suit.
Today, AI is the big wave reshaping the tech landscape.
Tech giants and startups alike are racing to innovate with AI, with hyperscalers from Google to Amazon looking to expand cloud AI solutions for various use cases. In 2023, 26% of US startup investments went toward AI companies — more than double the amount of investments than the year before.
Cloud skills have been increasingly important in tech, but will AI change that? The answer is no. It's important to note that AI is building atop the foundation of the cloud.
The growth of AI means we'll see more work set out for skilled cloud developers — in spite of layoffs affecting other sectors of the tech industry. In fact, the World Economic Forum estimated that AI has the potential to create 133 million new jobs by 2030.1
Today, I'll talk about how AI is changing the cloud, and what that means for the future of developer jobs.
The AI boom and the cloud #
2023 saw a mushrooming of AI solutions and adoption in various industries, with AI enhancing cloud solutions.
For one, AI-powered analytics are enabling more powerful tools. ML models can fortify cloud security by responding to threats in real-time, as we see with AWS Amazon GuardDuty. We're also seeing more powerful natural language processing tools. AWS, for instance, offers Amazon SageMaker, a solution leveraging AI to simplify machine learning model development.
AI-enhanced services offer benefits to consumers and businesses alike. AI automations are streamlining routine management tasks to create more efficient cloud environments. Models can analyze usage patterns to inform scaling needs for cost and resource optimization. Meanwhile, user experiences will get more personalized. Streaming platforms and e-commerce can make stronger recommendations, while search engines can leverage NLP to generate more relevant results for search queries.
Over the last year, top tech companies have been racing to develop AI solutions. Google launched their own Gen AI Chatbot: Google Bard. Oracle upgraded their Oracle Cloud Infrastructure (OCI), enabling training capacities that gained them $4 billion in contracts from AI development companies.2 Meanwhile, Microsoft has partnered with various organizations to offer AI solutions in the healthcare sector, such as the Azure AI Health Bot, which supports administrative, clinical, and patient care tasks.
Generative AI technologies have accounted for many of the new AI solutions. The figure below shows how individuals have used Generative AI in their workplace, according to a McKinsey & Company survey.3
By making off-premise computing possible, the cloud enabled this widespread use of AI. Without the cloud, most companies and individuals don't have the infrastructure and resources needed to run powerful AI tools.
That said, the cloud's infrastructure has met its own limitations.
AI necessitates new iterations of the cloud #
Despite their prevalence, cloud infrastructure wasn't built to handle large AI systems — and the industry is facing new challenges as a result. The majority of today's cloud servers run on general-purpose CPU chips, which are not able to sustain the heavy workloads needed for AI models.
To handle the complex computations needed for AI, hyperscalers have no choice but to optimize for AI or be left behind. Among them, Oracle and Google upgraded to GPU-enabled servers that use NVIDIA H100 AI processors. In August, Google Cloud unveiled their fifth-generation TPUs, which are optimized for high inference and training performance for LLMs and gen AI models. AWS is also working to optimize for AI — despite the fact that they already own over 50% of ARM-based servers (which were once competitive choices for ML and AI). Helping data centers optimize for AI is now its own profitable service, taken on by startups such as MangoBoost.
An AI-optimized infrastructure means there will be changes in best practices for building and scaling applications — as well as the challenges that come with designing them. There'll be new iterations of how we handle and manage data. Most organizations are aware that the rise in AI comes with new risks to address.
Notably, data accessibility and security will be a huge priority for these systems. Pipelines for preprocessing and clearing data will be essential for data quality and product success. Meanwhile, new security measures will need to be implemented. With LLMs parsing data, allowing access to the wrong data can have dangerous consequences. Consider enterprise data, for instance: Without the right restrictions in place, an employee could ask a chatbot a question and access to internal information they shouldn't have — from details about their performance evaluations to restricted financial data.
All of these changes mean that developers have lots of work to do to ensure these scalable systems fare well throughout AI's proliferation. Developers will have to rethink, restructure, and rebuild applications. As customers begin to expect more AI-driven insights and solutions, we'll need to optimize cloud architecture and server-side design.
AI + cloud = New opportunities #
Without a doubt, AI is ushering in a new era for more efficient and insightful cloud solutions. These new tools will boost our productivity, streamline operations, and be widely available thanks to the cloud.
For this shift to AI to be as successful as possible, the industry needs skilled cloud developers as we rethink, rebuild, and face the challenges unique to this new paradigm.
If you want to be among the developers contributing to the future of cloud/AI, you'll need to have a deeper understanding of cloud services and how they work.
At Educative, we offer 1,000+ Courses, Projects, and Skill Paths to help you upskill and keep up with the changes.
Our newest offering is CloudLabs, a setup-free way to get you hands-on skills with AWS services from SageMaker to Lambda. You can browse our growing catalog of 100+ CloudLabs here.
Here are a few popular CloudLabs that may interest you:
We also offer Courses and Skill Paths written by industry experts. To help you get practical, hands-on experience with in-demand skills, you'll code as you learn with an in-built coding environment.
Here are some that will equip you with cloud skills:
Cloud Native Development with Tailwind, Google Cloud and Firebase
Cloud Architecture: A Guide To Design & Architect Your Cloud
You can also try out Educative Premium with a free trial today. While the Premium tier doesn't include CloudLabs, you will get access to our Courses, Skill Paths, Projects, Personalized Paths, and more.
Remember, times of change are also times of opportunity. I hope you take this chance to level up your skills for the future of cloud services.
Happy learning!
Free Resources
