One of the key problems with being a tech worker today is that it is hard to know whether your job will be automated out of existence in the not-too-distant future. For those that work in tech that fear for their jobs, the lines between worry and fear can become very blurred as the constant deluge of information regarding the latest model, the new automation tool, and the impending demise of the human engineers of software to be replaced by automated software and programs is almost too much to bear at times, with the constant hum of the laptop at midnight feeling increasingly ominous.
While it is true that technology is changing from day to day and that new software and tools are constantly being created to assist the engineer, it is also true that there are fundamental issues that require human judgment, human empathy and a deep understanding of systems. The work of the architect and of the researcher requires a depth of knowledge and a level of sophistication that is not currently possible with the aid of software.
When it comes to protecting your future, it’s good to know where the human edge still exists, and can kick in.
The Systems Architect: Mapping the Unseen
Writing a single function or local error fixing can be automated more and more over time. However, building out a large, highly available and secure infrastructure that can be scaled while fulfilling different business goals is not.
An architect of systems does not only have to look at a piece of code to check whether it is correct. They have to look at a whole lot of empty space around the system, which is to say around the interfaces to other systems that it interacts with, and also to the wider environment in which it runs. It is very difficult for a human to even begin to understand the potential knock-on effects of a single database failure in a region somewhere. (By “potential knock-on effects,” I mean the effects on customers and so on elsewhere in the world of that database failure.) For instance, a failure of a database in one region might cause a series of failures in customer service in a completely different region of the world. It is the kind of situation in which things go wrong very fast, and in which there is a lot of pain. The systems architect has to try to understand this kind of situation and then to make a decision that trades off between the cost of the implementation of the fix for the problem and the potential technical debt that the organization is going to incur as a result of that fix over the next five years. There is no doubt that there are many architects who wish that they had more automation to help them in these kinds of decisions. But at the end of the day, the human is required to make the hard choices. A system can certainly map out a map for an architect to follow, but that architect must then decide where to build a road through a shifting terrain.
This work requires understanding of risk as well as the organizational culture and its ability to withstand long-term costs. Architects make choices, and there is often no “best” answer.
A machine can optimize a map for routing. But a Systems Architect must create the map in the map for routing. When the map itself is changing all the time (e.g. because of new buildings or new roads being added all the time), then there is no existing map for a machine to optimize. A Systems Architect must figure out where to put in the road in the first place.
The Product Manager: Deciphering Human Desire
Product managers have a very uncertain day as a product manager. Unlike architects who work with clean data to create blueprints for systems to be built, and many engineers who work with clean data to build software, a product manager’s work is typically filled with lots of ambiguity, messy human behavior, conflicting points of view, and politics within a company.
Also important is that when a user asks for a feature, they actually are asking for something else. Deciphering the core of the frustration of the user is very important for the success of a product, and the architect of that product needs to have a lot of empathy for the user and listen a lot to him. As mentioned before, deciphering the core of the frustration of a user is not very easy as sometimes the user does not even realize what is frustrating him, and it is very important for the Product Manager to have a lot of patience and to be able to negotiate.
So, this is a job of negotiation, of trying to get people to do what you want. And that involves getting engineers to write code, to design and implement a feature. And then you have to get senior managers to authorize money to implement the feature. And then you have to get a design team to create an interface for the feature that is usable by users. And all the way through, you are trying to find some sort of compromise that satisfies all of the various parties involved.
Cybersecurity Incident Responders: Managing the Chaos
Incident Responders have to deal with two things: the first, they have to solve a technical problem (which might be hard enough already), but secondly, they have to deal with a human crisis as well. The technical problem might be solved within hours or days, but the human crisis can last for weeks or even months. And it’s exactly this human crisis that makes their job so hard.
He notes that, “responders must manage to deal with incomplete information, compromised systems, and demanding stakeholders during the worst moment of their lives.” To make such split-second decisions, cybersecurity professionals must have “years of experience with worst-case scenarios.” And, of course, that means working long hours under intense pressure.
Security professionals have to think like adversaries. Often, they can’t even anticipate what kind of creative choices a human hacker might make. And then there are the rapid ethical decisions that must be made while containing a threat. For example, a security incident responder might have to take a critical system down for thousands of users in order to contain a threat. It’s a huge burden to put on someone to make such decisions in a split second.
It’s a weighty responsibility that code cannot undertake. This is to find, to interpret, to manage the problems in the very dynamic and constantly-changing terrain of information itself, to make the decisions, and to take the actions to manage the huge numbers of people affected by these.
The UX Researcher: Finding the Friction
UI Design focuses on surface level of abstraction, i.e. rules of visual layout. User Experience (UX) Researchers, on the other hand, study real people interacting with technology.
Researcher work on UX has a very human twist – they sit down with real users and study their behavior while they use a product or service. Every pause, every sound of frustration, every time a user seems to be confused but is too proud to admit it is all noted down. Analyzing such human psychology requires an understanding of many different factors: users’ cultural background, their current emotional state, and the many different aspects of accessibility.
But to understand human psychology, there is also a need to understand the cultural background of users and their emotional states while they are interacting with technology and also their accessibility needs. So there is a big difference between user testing a financial service on the subway and in an office. While a researcher can analyze data from users interacting with technology, they will not be able to fully understand the causes of human frustration unless they can relate to it themselves. Why would we expect an algorithm to understand and deal with human frustration when the people who developed it have not themselves experienced that frustration?
AI Safety and Ethics Engineer: Building the Safest AI.
But first, the “fun” job of keeping all that automation under control – and ensuring that it behaves in a safe and fair manner in the real world. While building very powerful software is child’s play for many, keeping it safe and fair for users and customers is an enormous challenge.
AI safety engineers don’t just look at technical accuracy. They look at systemic bias, historical inequities, and the societal impact of automated decisions. They must ask difficult questions: Who does this algorithm leave out? What are the unintended consequences of optimizing for this specific metric?
No. While the work will often begin with data analysis and simulation, a large part of this work is in determining what is ‘fair’ or ‘ethical’. Such matters are philosophical, and, indeed, cultural and legal. All of these are deeply human issues. So this work is not amenable to the usual approach of Automation and therefore must be done by humans. But it is amenable to humans working together as a team to find the solutions.
Navigating the Transition: Updating Your Credentials
It can be helpful to reframe your resume to better highlight your strengths in high-level technical positions, such as systems design or product management. This will be especially important for showing the employer the human-centric solutions you have developed under pressure, as well as demonstrating your ability to work in highly ambiguous circumstances.
To give your resume a fresh look, explore Zety’s plans and pricing. Their platform allows you to choose from a variety of templates that easily allow you to highlight your project management experience, alongside your technical and soft skills, to help you stand out to potential employers.
Refreshing your professional profile takes more than simply listing out your technical skills in a long list of programming languages. The best way to demonstrate your proficiency in the many different roles technology plays in serving people is by highlighting your experience in specific project management-type roles and listing out your various soft skills in addition to the technical skills that you possess as a developer or engineer.
Looking Ahead
Technology is always evolving. Just as the tools you use at work today are likely to change in the future, the tasks that you perform on a daily basis will change as well. The more that your work involves writing code, the less that will be the case in the future. In place of writing code, your work will consist of tasks that require critical thinking, communication, and an ability to understand the needs of people. All of the roles highlighted above will continue to evolve, but are likely to remain at the core of a technology company. This makes them perfect roles for people looking to build a career in technology that will be resilient in the future.
