The “Good Engineer” Checklist

As shown in Figure 4-1, aside from the aforementioned core pillars that define a good professional, good software engineers are

• Action biased instead of overplanners.

• Iterators. They understand the importance of iterative development and realize that as long as a contribution improves the system, that contribution does not need to be perfect. Continuous iteration results in constant improvements.

• Capable of breaking down larger concepts/problems into smaller ones.

Four major factors of a good software engineer illustrate good analytical skills, anticipates problems, technical knowledge, humility and self-confidence, and good communication.

The “Bad Engineer” Checklist

A bad engineer is basically the inverse of the previous two sections. Here is a list of behaviors commonly found in underperforming professionals. These behaviors are the direct consequence of a lack of the attributes described in the previous section and are summarized in Figure 4-1.

Keep in mind that the points listed above are behaviors that bad engineers do consistently, great engineers can have a bad day and slip into those behaviors occasionally but they certainly shouldn’t be a habit for them.

Five major factors an engineer illustrates cannot justify decisions, failing to interact, not being proactive, being too uncooperative, ignoring feedback, and being unable to give insight.

The Importance of Experience

In slightly more complex terms: experience allows us to deal with uncertainty. Most importantly, “unknown unknowns.” That is, it enables us to deal with the correlation between experience and a broader knowledge that decreases the surface of things that we do not even know we don’t know (unknown unknowns). In his book “Risk Intelligence,” Dr. Dylan Evans illustrates this notion (depicted in Figure 4-3) using the concept of a darkened room:

“Picture your mind as a lightbulb shining in an otherwise dark room. Some nearby objects are fully illuminated; you can see them in every detail, present and identifiable. They are the things you know very well: the names of your friends, what you had for breakfast this morning, how many sides a triangle has, and so on. The objects on the other side of the room are completely shrouded in darkness. They are the things about which you know nothing: the five thousandth digit of pi, the composition of dark matter, King Nebuchadnezzar’s favorite color. Between light and darkness, however, lies a gray area in which the level of illumination gradually shades away. In this twilight zone, the objects are not fully illuminated, but neither are they completely invisible.”

Assuming that an individual learns from both their past experiences and the experiences of others, they will, with time, become better at navigating this twilight zone. That is what we mean when we say that “experience allows professionals to deal with uncertainty.” Note that this does not imply that the twilight zone will shrink with experience. Of course it might if the individual remains in the same domain for years and gathers more and more minute knowledge. But in most cases, the twilight zone will always remain—people switch jobs, tackle new challenges, or venture into new domains. With experience however, the individuals become more self-aware of their limitations, the limitations of their field as well as how much relevant information they may not have to make a decision. Over and under-confidence are replaced with better self-knowledge, and the individual becomes “better calibrated” to their task environment. Not only does this allow for more effective task execution as risks and uncertainties can be identified more readily, it also makes for better communication: The more experienced someone is in a field, the better their educated guesses and probability estimates become.

In contrast, inexperienced professionals might have learned the necessary technical skills to execute a task that lies “outside the twilight zone,” but they usually are unable to make (or even identify) far-reaching decisions and voice them with the right amount of certainty. Without having been bitten by their own mistakes a sufficient number of times, inexperience means bringing either a “dangerous excess or a problematic deficiency”⁶ to the table. An inexperienced candidate has not had the time to make all the mistakes that they need to make. Since they do not know what these mistakes are (or how to avoid them), chances are that they will make at least some while working on your project.

It is therefore no surprise that experience is highly valued in the field of software engineering. Engineers can make multiple important decisions in a short time frame that will deeply impact the future of an entire project, especially in the initial phases of a project. Therefore having a good understanding of the consequences of those decisions in the future, the knowns, unknowns, and unknown unknowns, will directly impact the success of the project.

To put this all into more concrete terms: There are decisions that are easy to change later, like choosing a certain existing library for a very specific feature. However, other types of decisions, such as selecting a programming language or framework for a project, are much harder to change or revert. Experience plays an important role in this context, as it not only allows the engineer to properly differentiate between the different types of decisions, but also enables the engineer to handle difficult decisions more effectively. That is, when making a decision that is difficult to revert in the future, they can take the necessary steps that could make a potential reversal less difficult. For example, applying different techniques to properly isolate code and hence minimizing the impact of future changes.

Decision making and uncertainty aside, experience also acts as a natural filter for unsuitable job candidates. As Tim Nichols concisely puts it in “The Death of Expertise”:⁷

“Every field has its trials by fire, and not everyone survives them, which is why experience and longevity in a particular area or profession are reasonable markers of expertise.”

A circle diagram of three factors. The Unknown unknown comprises the known unknown, and the known.

The Dunning-Kruger Effect

We probably all know this one incompetent colleague who was consistently unable to deliver, but thought the world of themselves. It turns out that there is a name for this type of illusory superiority: the Dunning-Kruger effect. First proposed in the late 1990s by two psychologists, David Dunning and Justin Kruger,⁸ the effect basically states that the incompetent are unaware of their incompetences due to the lack of competence (Figure 4-4). Put more bluntly: the stupid are too stupid to realize that they are stupid. For example, an incompetent software engineer may be unaware of their inability to design and write good software, since the very skills that would allow them to develop such software are the same skills that allow them to identify “bad” software. The irony therefore follows that, in order to make a bad software engineer realize their incompetence, they must become more competent.

While, according to Dunning and Kruger, the incompetent are overconfident in their abilities, the more competent a person becomes, the more underconfident they become, as through knowledge they become blissfully aware of the limits of their knowledge. It is therefore no surprise that many companies profess “humility” as one of the core values that they look for in candidates. Sure, the “overconfident know it all” will be an annoying colleague and make for a bad culturefit. But at the heart of the boasting and bold candidates often lies a lack of knowledge and competence. Attributes such as “humility” and “modesty,” on the other hand, are often a signal for competence and should be actively screened for when hiring.

A chart depicts the relationship between field confidence and knowledge. The curve begins at a low point, rises at a constant rate to a high point, and then moves to the downside.

Honesty

Finding new hires is expensive (as we will see in the next section “The Cost of Hiring”). To ensure that these expenditures pay off (that is, to ensure that you actually retain your new hires), honesty during the interview process is crucial. Expecting job applicants to talk honestly and openly about their past experiences is a given. Unfortunately, many employers do not see honesty as a two-way street. Sugar-coating the position and company is common practice, either because the interviewer does not want to sound too negative (and hence be responsible for painting the company in a bad light) or because they are afraid of failing to attract the right candidate. This is problematic in two ways. Firstly, by lying, you are acting unethically. As Sam Harris eloquently puts it in his book “Lying,” when he writes that “[by lying] we deny others a view of the world as it is. Our dishonesty not only influences the choices they make, it often determines the choices they can make—and in ways we cannot always predict. Every lie is a direct assault upon the autonomy of those we lie to.”

That is, by sugar-coating a position, you deny the candidate to see what they are actually signing up for. Since people spend most of our waking hours at work, a candidate’s understanding of the company will have a significant impact on their life if this understanding turns out to be false.

Secondly, by lying to candidates, you will end up generating additional costs for the company, as honesty and expectations are closely linked. You are hiring intelligent, capable candidates. And—surprise surprise—smart candidates will soon figure out what is wrong with the company. You, in turn, will gain a reputation as a dishonest leader, and will lose the trust of those working for you. A lack of trust combined in a company that fails to meet expectations results in brain drain: If the job market permits, people will leave. Which means you will need to start the hiring cycle all over again.

One prevailing source of dishonesty in leaders comes from unfeasible targets. An honest leader, operating in an honest environment, will never purposely aim at unfeasible targets. Leaders that catch themselves thinking about a $50-100 billion valuation in 5 years while the company only has 500 users should ask themselves how honest they are with themselves. While your job as a leader requires you to remain positive even when in adverse circumstances, you should remember that there is a balance to be struck between positivity and realism. Especially in business, honesty and realism go hand in hand. As Ayn Rand once famously said: “You can ignore reality, but you can’t ignore the consequences of ignoring reality.”

Another common source of dishonesty stems from sugar-coating challenges. Leaders can be upfront about the technical and organizational challenges that await candidates without needing to put everything into a completely negative light. Unfortunately, most leaders prefer to try and make reality look better than it really is.

Summarizing these two major sources of dishonesty in interviews: If you can be honest about the status quo and what the company is aiming at, then you will be able to avoid a significant percentage of churn.

The Cost of Hiring

Hiring involves direct and indirect costs. The direct costs, such as man-hours spent by recruiters or advertising, are obvious and as such we won’t elaborate on those here. The indirect costs however are more subtle, yet usually greater. Recruiters naturally like to push as many candidates through the pipeline as possible. This approach leads to lost engineering man-hours. Technical interviews need extensive preparation, and technical exercises require reviewing. By narrowing the pipeline—that is, detailing the exact job requirements as best as possible, as well as setting high screening standards—a company can reduce these costs.

Once a candidate is hired, they will need to be on-boarded. Not only is this engineering time that is spent not developing the product (opportunity cost), but the on-boarding phase is likely to introduce risks to the product. New bugs tend to be introduced as new engineers begin working in a new, unfamiliar environment. While this decreases over time, and good engineers will more than make up for the cost of hiring (otherwise they would not be hired), it is important to keep this cost in mind as companies with a high churn will pay a high price.

The cost of hiring increases the higher your churn is. Once you begin losing a certain percentage of your teams, you not only lose knowledge but also impact team morale and team processes. Even if you have perfect hand-over processes and perfect documentation, lost knowledge is difficult to get back and usually takes many months (or years). A loss in morale on the other hand can result in a snow-ball effect: colleagues leave and either bring others with them to the new company or motivate them directly or indirectly to jump ship.

Last but not least, processes die with people. People enter a company and bring with them new processes and establish these for different reasons, experienced professionals implement them knowing their pros and cons, while inexperienced ones do so because they are comfortable with it and want to mitigate changes. Hopefully, the new processes might make actual sense and improve things. However, many times they are introduced because people want to set a mark, be seen to be doing things, or simply because they are used to working this way. Regardless of the “why,” the introduction of new processes by new hires can often fan an existing fire: People are creatures of habit and like working in ways that they are used to. A large influx of new hires and new processes can therefore increase dissatisfaction and result in further brain drain.

The Cost of Bad Engineers

Before we begin to discuss the price that a company pays for hiring “bad” engineers, we need to distinguish a bad engineer from a bad hire. On the surface, this difference might not be immediately obvious. Nevertheless, it is an important one to keep in mind, and the essence of it is this: Sometimes a hire might not be the ideal fit for a team/company, but could still be a good fit for another team/company. Every single one of us is subject to situational forces, meaning that our surroundings define and exert certain processes of transformation on our character, changing how we act and react to those around us. Put into one team, a certain engineer might excel at tasks that, if placed in a different team, they would perform poorly. Being able to differentiate the causes of poor performance—that is, poor performance due to their surroundings, or poor performance due to incompetence—is difficult and requires careful observation. The cost of bad engineers is therefore different to the costs associated with bad hires. The latter can often be rectified or compensated with relatively little change and poses a lower risk to the company as a whole. This is because bad fits are due to situational forces, while bad professionals are “bad” because they lack certain attitudes, abilities, skills, or intelligences that are required in order to excel at the job regardless of the team.

Above all, bad engineers that are allowed to persist in their positions will set bad precedents and lower the standard of the team as a whole. Most people tend to sink to their lowest acceptable levels of performance. By not removing unsuitable team members, leaders set a low bar and signal to others that poor performance is acceptable. Once settled, these behavioral and cultural habits can be difficult to reverse even if the responsible individuals have been removed.

Retention: Understanding the Career Cycle

By now we have hopefully convinced the reader that hiring is expensive. One important factor in reducing the cost of hiring is retention. That is, keeping good employees for as long as possible. Based on our experience, at the heart of retention, lies the understanding of the cyclical nature of careers (depicted in Figure 4-5), meaning that we live out our careers in cycles. Depending on the individual, the length of an individual cycle may vary—some people’s careers consist of many cycles; other people’s careers consist of few cycles; and some very rare individuals might only live through one cycle. Furthermore, the shape of the individual cycles themselves may differ greatly.

A parabolic shape curve indicates the relation between productivity and time. The curve starts from onboarding and reaches the maximum peak contribution point and falls down.

When we first join a company, our career enters the “on-boarding” phase: We do not yet know much about the technicals of the project we work on, the processes and internal workings of the company itself. As such, we are in a strong learning phase, and are not yet very productive. As time goes on, we become more familiar with our task environment and become more productive. We are enthusiastic, are constantly learning and discovering new things, and are trying to maximize our productivity and output. During the honeymoon phase, we see nothing “wrong” with our surroundings. We are dazzled by all the nice people around us and the world is our oyster—career opportunities glisten before us, and we put all our heart and energy into the job. We are at “peak contribution,” meaning that we have learned a sufficient amount to be highly productive. As time goes on however, we start to see the cracks in the walls: politics, colleagues that we do not get on well with, a difficult boss, some big-picture decisions that we do not agree with, better pay elsewhere …Our enthusiasm wanes. Slowly we become used to the rut, and execute with less energy and enthusiasm. Stress might consume our remaining energy, and there isn’t much to learn. We have entered the phase of decline. At the end of this phase, we either transition into a new role within the company (a promotion, a new team, or a new department) or we leave the company. Regardless of our decision, the cycle starts again.

The key to retention is therefore this: understand the cycle and realize when employees enter the phase of decline. Chances are you will never be able to prevent the decline phase (the world of work is too complex for this), but you will be able to stretch out the productivity and honeymoon phase a lot (see Figure 4.6 for examples of short and stretched out honeymoon phases).

The relationship between productivity and time is illustrated by two parabolic half-heart graphs. The first graph depicts a rising curve that reaches a maximum and then falls. The second graph represents the inverse.

A semi-arc curve illustrates the relationship between productivity and time. The curve begins with onboarding, rises to its maximum peak contribution, and then falls.

The ideal cycle on-boards quickly and stretches out the honeymoon and peak phase for as long as possible, hence maximizing retention. Employees whose careers follow this shape remain productive in their positions for a long time, and only decline eventually.

Practical Tips

Hiring good engineers is easier said than done. And so far, we have done the easy part: We discussed hiring and retention in theoretical terms. Now comes the hard part: putting the theory into practice, and defining a set of concrete guidelines that you can apply immediately.

Conclusion

Everybody knows that, if you want your company to grow, you need to hire good professionals. However, many people forget that hiring is not just about growth: it is also about retention and maintaining a solid baseline. A few bad hires can destroy entire teams. To complicate things further, finding good talent is difficult. As such, we outlined a set of attributes that will help you hire the right candidates. In summary, these are humility, experience, a high degree of self-consciousness, technical ability, good analytical and good technical skills as well as the ability to anticipate problems before they happen.