At AWS re:Invent 2021, AWS made further inroads into the no-code ecosystem with the launch of Amazon SageMaker Canvas. SageMaker Canvas combines the highly technical discipline of machine learning with the low barrier to entry of no-code. AWS surely hopes this broadens its customer base with both no-code advocates as well as customers who lack dedicated data science teams.
But what are the longer term implications?
Accelerate your career
Get started with ACG and transform your career with courses and real hands-on labs in AWS, Microsoft Azure, Google Cloud, and beyond.
From the perspective of machine learning, making such powerful technology available in a “black box” format means users either will not (or will be prevented from) understanding what’s going on under the hood. In the context of no-code, there’s potentially an entire population of cloud infrastructure engineers left to wonder if their skillset suddenly has a shelf life.
First, let’s jump in and look at what no-code is (and isn’t).
What is no-code?
No-code is a newer paradigm in application software development. No-code, as the name implies, eschews coding/programming to create logic, instead offering graphical and interactive interfaces that allow users to create fully fledged software applications with minimal configuration and technical acumen.
Although no-code has seen an explosion in popularity in the last two years, long-time denizens of the internet are likely familiar with precursor services like IFTTT, which offer modular, connectable automation logic through a simple user interface.
Fast forward to the present day, and tools like Integromat and Zapier offer powerful automation suites, while services like Webflow and Bubble enable users to create fully functioning interactive websites using a visual interface.
The no-code boom makes perfect sense: taking an idea and turning it into functioning software requires either knowing how to code, or partnering with a developer to create an MVP. Now, anyone can quickly turn an idea into a working application prototype.
In our current fast-paced on-demand era, the ability to quickly iterate on and deliver ideas is a boon for would-be entrepreneurs and creators.
No-code obviously empowers small teams and solo creators, but how it slots in with the AWS service offerings is an interesting question.
How does no-code fit in with AWS’ long-term strategy?
SageMaker Canvas is actually not the first no-code service to be offered by AWS.
Last year AWS launched Amazon Honeycode, a visual web and mobile application builder that operates on the spreadsheet model popularized by such services as Airtable. This year, AWS also launched AWS Amplify Studio, a more comprehensive no-code/low-code visual development service that allows users to create holistic application stacks, including both frontend and backend components.
Although AWS has simplified many of the technical aspects of delivering highly scalable highly performant applications, most users will tell you it still requires a fair degree of competency, and complex AWS architectures are often managed by teams of multiple engineers.
So why is AWS choosing to offer tools that seem to not require engineers at all?
The simple answer to all this is a term that actually originates from the sales and marketing side of the house: Total Addressable Market (TAM). TAM basically defines the superset of all potential paying customers of a given product or service. The bigger the TAM, the bigger the potential revenue.
Any company generally wants a large TAM for their product or service, and in the case of AWS they want it as big as possible. However, AWS (and other cloud providers) are faced with a skills-gap problem.
The demand for cloud-focused technical skills is so great that many roles are left unfilled for months. That may sound like a good thing, but customers who cannot fill critical roles are likely to delay or re-think cloud-based projects, cutting into the aforementioned TAM. The gap is so big that Amazon launched an initiative with the goal of training 29 million people in critical cloud technology by 2025.
Creating a new generation of cloud-capable engineers isn’t something that can just happen overnight though, and AWS product teams want to drive revenue now and in the future. The no-code ecosystem fits in well with a strategy to expand TAM; AWS can tap an entire demographic of non-technical users who might have otherwise not considered AWS and a traditional cloud offering.
With the addition of machine learning tools, customers who don’t have dedicated data science or data engineering capabilities can now empower their marketing and analytics teams with machine learning capabilities. Ideally, as these customers scale, they begin to take advantage of other AWS services as well, increasing AWS revenue.
But what about everyone who already is an engineer? What happens when AWS launches a no-code distributed systems tool? Or a no-code NoSQL database? Are those skills going to be obsolete? Another question to consider specifically in the context of machine learning and AI: what are the potential ethical pitfalls of opening up this kind of capability to inexperienced users?
Watch: Recapping re:Invent 2021
Who were the big winners and losers out of re:Invent? In this free, on-demand webinar, our panel of AWS Heroes review all the biggest announcements and chat what it means for the year ahead.
Should engineers be worried? Should everyone?
If there’s one constant in the technology industry, it’s change. Skills that were once a guarantee of a rewarding career (Adobe Flash, anyone?) have all but disappeared, while some huge technologies aren’t even a decade old (React).
If change is constant, then no-code is just another evolution in continuing technical progress. That doesn’t mean present day engineers should cede their jobs to inevitability. Nor does it mean we as a technical community should not engage with and understand the capabilities of these new services.
It wouldn’t be a stretch to assume a decent number of cloud engineers have backgrounds in more traditional, legacy IT roles. When the cloud took hold, these engineers needed to adapt and learn a new set of skills. No-code will not be any different.
Even for infrastructure and systems-focused engineers, no-code offers several potential advantages. Basic automation stacks, even the simplest of which typically require a few hours of work, testing, and support, can now be designed and deployed easily.
Developers and non-technical users no longer need to depend on infrastructure teams for provisioning simple automation, freeing them up to work on more interesting and challenging technical problems of scale. Traditional code-based architecture could potentially be combined with no-code applications to create hybrid systems that are accessible to both technical and non-technical teams.
No-code and machine learning adds another wrinkle to the discussion. While it democratizes a domain once reserved for only the most technical of practitioners, machine learning and AI technology has given rise to a multitude of ethical concerns:
Ben Kehoe — a Cloud Robotics Research Scientist at iRobot and an AWS Serverless Hero — ostensibly raises a valid concern. In a technology that already sees only very limited and highly specialized usage, what are the implications of making it “easy” for everyone to use? What are the responsibilities of the users, and what are the responsibilities of the provider?
A very interesting question for the future is whether or not AWS will update it’s Shared Responsibility Model to include responsible usage and deployment of machine learning and artificial intelligence technology.
No-code and the future
No-code is likely going to be a fixture in technology for the foreseeable future, both as a general development platform, and specifically as part of the ever-growing AWS product and services portfolio.
It will be interesting to see how AWS intends to grow its no-code offerings, and how that potentially impacts the matrix of desirable skills for a modern engineer/technologist.
Like previous generations of IT workers, those who adapt and grow will have better positioning for a new wave of roles, some of which may not even exist yet. Those that choose not to may find themselves ultimately left behind.
The implications of no-code machine learning/AI seem less clear, and everyone in technology can and should pay close attention. Ethical concerns aside, the more time spent by people building even rudimentary solutions will certainly advance the field, and that has broader implications for technology and society as a whole.
Other AWS re:Invent 2021 highlights
- AWS re:Invent 2021: The biggest announcements
- S3 Glacier Instant Retrieval deep dive: Which S3 Storage Class is right for me?
- AWS aims to reduce cloud barriers, expand industry-specific focus
- AWS just dropped a game changer for startups, small business
- Machine learning just got more accessible and inclusive at re:Invent 2021
About the Author
Mike Vanbuskirk is a Lead DevOps engineer and technical content creator. He’s worked with some of the largest cloud, e-commerce, and CDN platforms in the world. His current focus is cloud-first architecture and serverless infrastructure.
Get the Cloud Dictionary of Pain
Speaking cloud doesn’t have to be hard. We analyzed millions of responses to ID the top terms and concepts that trip students up. In this cloud guide, you’ll find succinct definitions of some of the most painful cloud terms.