We started Acclimate to work collectively towards an AI-enabled society. Why, you might ask, would we want to live in a world filled with artificial intelligence? AI, specifically machine learning, has the power to fundamentally transform society by making everything frictionless and by giving people more fulfilling lives. This post won’t go too deep into why AI will have this effect (there are plenty of articles on the subject), but rather the reasons why we are spending our time helping businesses design and implement data-centric strategies and capabilities.

The timing is right

"We have enough papers. Stop publishing, and start transforming people’s lives with technology!" – Andrew Ng

The world is finally overcoming the disillusionment that has surrounded artificial intelligence (AI) for the last six decades. Machine learning (ML), a branch of AI that enables computers to “learn” from data, is more useful than ever before. Cheaper, faster, smaller, and more efficient sensors, processors, storage, and communication protocols have dramatically increased the power of ML algorithms. Specifically, deep learning has emerged as a dominant force, giving machines human-like perception in vision and language . Businesses across all industries are beginning to understand the power of these new capabilities, and the ecosystem for building them is quickly maturing.

We have at our disposal an increasing number of open source algorithms and tools that anyone can use. Cloud providers offer access to world class hardware to train and run machine learning models. These services can easily be scaled up or down and can be pieced together over time as an ML system evolves. The sheer number of technologies is daunting, but all the components are there. Connecting it all together to drive real value is the challenge, which is why we need machine learning practitioners.

AI creates massive value

"We’ve been seeing specialized AI in every aspect of our lives, from medicine and transportation to how electricity is distributed, and it promises to create a vastly more productive and efficient economy" – Barack Obama

AI researchers have always been excited about building an artificial general intelligence—an AI system that could perform any task a human can. The implications of such a system would be huge, but at the moment this panacea is a work in progress. In the meantime, we are finding use cases for narrow AI all over the place. All industries will be affected by machine learning, but some experience the effects sooner than others. Transportation and adjacent industries are radically changing as self-driving cars and trucks hit the road. Doctors’ capabilities are being augmented with hyper-accurate diagnostic tools and patients are receiving customized treatments and monitoring. Finance is becoming increasingly automated with intelligent fraud detection and robo-advisors balancing your investment portfolio. The retail industry is segmenting customer groups, providing tailored recommendations, and extracting insights about product usage.

This is only the tip of the iceberg. Machine learning algorithms will permeate the fabric of society. All of our resources (energy, water, raw materials, compute power) will be generated, stored, and distributed more efficiently. Humans will make better decisions using data-driven insights; many decisions will be made automatically without any human oversight. We will interface using only our voices, and computers will fade into the background as they better predict our needs. Everything in society will become frictionless and people will have time to pursue more fulfilling work. Ok, this is getting a bit futurist, but you get the point. There are plenty of immediate uses for machine learning we should focus on, but in the long run it will fundamentally change the way we live our lives.

Implementing AI is hard

"Big Data is like teenage sex: everyone talks about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so everyone claims they are doing it." – Dan Ariely

AI is facing challenges on both the demand side and the supply side. Incredible, untapped value is hidden away because those who need it most don’t yet know what it can do for them. Secondly, the scarce talent to unlock this potential is being seized by big tech companies and agile startups. Let’s address these two issues (demand and supply of AI skills) one at a time.

As you understand, AI can provide businesses a strong competitive advantage. Not implementing a data strategy means that it will become increasingly difficult to compete with those who do. Using data and AI to make informed decisions, increase operational efficiency, and provide an excellent customer experience is more critical than ever. The challenge is understanding how AI can be used to achieve these outcomes, and then designing and building tailored solutions. This all begins by asking the right tactical and technical questions: Where is AI creating threats and opportunities in our industry? How is AI changing business models? What is our long-term AI strategy? Are we collecting the right data? Do we have a specific problem that AI can solve and can we build a prototype? How will this fit in with our existing architecture, and how will it influence people in the organization? Do I have the talent to execute my AI vision; where can I find it?

Hiring AI talent is challenging and expensive. A proficient ML practitioner would need to understand how businesses create and capture value, how to design and build AI systems, and how to seamlessly integrate and communicate with organizations. Building an ML application requires a multitude of skills: coding (R, Python, JavaScript, and many libraries), math (linear algebra, statistics, calculus), system architecture (cloud services, data pipelines, software design quality), and so much more. It’s not easy to find people who have a mastery of all these skills, who keep up with new research and tools, and who can apply all of it to functioning AI systems in the real world.

It’s important to realize that different people are required at different stages in ML development process. In the beginning, a business will need strategic, technical generalists who can see the big picture and build an end-to-end ML system. Only then can it be iterated on, and over time the business will need to bring on highly skilled specialists in data engineering, data science, data analytics, and app development. Hiring full-time specialists in the early days would be a costly mistake when what’s actually needed is a hacker with some business acumen and good design sense.

Businesses need to adapt

"If you need a machine and don’t buy it, then you will ultimately find that you have paid for it and don’t have it." – Henry Ford

For a business, investing early in a data-centric system that will support future AI needs will pay dividends. Not only is this desirable, but it’s necessary in this age of technological progress. The core data infrastructure is often the hardest thing to get right, but it is needed to support future capabilities. All of the components are freely available, but there is a lack of talent to connect them together in a way that will be agile and scalable in the long term. How then, does a business even get started building its data and AI competency?

We started Acclimate to solve this single challenge: getting businesses up and running with a functional data pipeline and a valuable machine learning application. We believe that the best first step is hiring a small, experienced team of ML experts who can flexibly scale up or down. We can help businesses understand how AI will affect them, identify the best immediate use cases for AI, as well as the long term strategic vision; design, build, test, and support a full-stack ML system; and build a data science team by finding external talent or training current employees.

We want to make an impact

"Artificial Intelligence, deep learning, machine learning — whatever you’re doing if you don’t understand it — learn it. Because otherwise you’re going to be a dinosaur within 3 years." – Mark Cuban

We have a few decades in our careers to create as much value as possible and to impact the world in some meaningful way. Why not pick a craft that’s difficult, massively needed, and endlessly interesting? We want to be on the forefront of this AI movement, staying on the cutting-edge technology-wise, but deeply understanding the business and societal implications. We’re dedicating our careers to it, and will continue to be harbingers of the change to come. We believe that AI technology will be transformative in our lifetimes, but that a select few will disproportionately benefit. It’s important for us to wield these tools responsibly, and to ensure that everyone else can too.

We’ve learned a lot, and now it’s time to apply that knowledge to help businesses build the capabilities they need to survive in this new environment. Our team comes together with a shared history to create a strong dynamic, with the ability to understand organizational needs and build full-stack intelligent applications. We take pride in our ability to communicate effectively and to build functional data products. This is where we can create the biggest impact in our careers, so let’s get to it.