Principles of Engineering Management

I consumed a lot of blog posts like this before becoming a manager. It feels satisfying to read vague principles like this:<p>&gt; Facilitate wellbeing<p>&gt; Personal safety, dignity, and wellbeing of every team member are paramount. Team success is only success if team members feel good about it.<p>But then you become an actual manager and realize that these are largely just feel-good aphorisms that don&#x27;t really help navigate the hard work of being a manager.<p>Don&#x27;t get me wrong: I think &quot;facilitate wellbeing&quot; and similar tidbits are valid advice. Likewise, a manager who chooses to <i>not</i> facilitate wellbeing would clearly be in the wrong. But when you get into management you realize that the hard work isn&#x27;t waking up and deciding to facilitate wellbeing. The hard work often involves making unpopular decisions, or telling someone &quot;no&quot; when their pet request wouldn&#x27;t be in the best interests of the team, or defusing interpersonal conflict among two team members who have been feuding for months.<p>These feel-good blog posts seem rather vacuous after being in the trenches of management for several years. Again, not because the advice is <i>bad</i> or <i>wrong</i>, but because it&#x27;s just a lot of text around the basics of being a decent human being. Learning how to do the hard things about management or how to make unpopular but necessary decisions is something that you won&#x27;t learn from the average blog post or LinkedIn post, largely because these function more as personal branding pieces than actual advice.<p>In my experience, the best management advice comes from talking to the best managers you know personally. Online, the most practical management advice comes on smaller forums where people are either anonymous (and therefore less interested in personal brand building) or older and accomplished (and therefore not worried about blowback impacting their career). The type of feel-good blog posts or LinkedIn blurbs that people associate with their personal brand don&#x27;t really add much value after you&#x27;ve read 5-10 articles saying the exact same aphorisms.

Lot of misconceptions here that cause teams to underperform.<p>1. Managing comes first. (Nope, as a manager your are primarily responsible for ensuring a good outcome. As your title goes from manager, senior, higher to VP etc this becomes more important. Lower job titles you get paid to &quot;do the work&quot;, higher titles are about &quot;achieve outcome&quot;.) In the current market you&#x27;ll often run into team members that don&#x27;t perform. I see this so often where leads think that they have to &quot;manage&quot; and not do the work and the result is them and their team failing 2. Facilitate wellbeing. That&#x27;s nice. But your primary responsibility is to A. Achieve the goal. B. hold people accountable to an acceptable level of performance<p>Give credit, take blame: That&#x27;s a good one.<p>This post is just full of vague language that doesn&#x27;t hold anyone accountable. Sucks to be an IC on that team, thinking everything is going well and eventually end up having whole teams being fired since your manager didn&#x27;t set the right pace.

&gt; Optimize work distribution &gt; Managers have a portfolio of work that the business needs and people with work preferences. Optimize the dual objectives of delivering value to the organization and giving individuals problems that build their skillset, impact, satisfaction, and&#x2F;or advancement. Performance is contextual; set people up to shine.<p>I find this so important. A good manager has a good mental map of the business&#x27;s short-term and long-term goals, and a corresponding map of each person on the team&#x27;s short-term and long-term goals (and strengths&#x2F;weaknesses) . This map requires constant updating and refining. The magic happens in finding the right fit between all of that.

This sounds pretty good, all these are qualities anyone would want in their manager if they had a job doing &#x27;the critical path of execution&#x27;. There&#x27;s few things more frustrating than continually having to halt in the middle of one task to jump over to another task because someone&#x27;s pants are on fire. Sure it happens, but it should be the exception not the norm.<p>However, there&#x27;s another side to the manager&#x27;s job that doesn&#x27;t seem to be addressed - interfacing upwards with whatever layers link to owners, founders, the board, shareholders, etc. How does that exactly work out in practice? Let&#x27;s say leadership makes what you think are really stupid decisions with disastrous longterm consequences (ex: Boeing 737 MAX design process, Google signing up with China to build Dragonfly, etc.). What do you do? Pour gasoline on yourself and set yourself on fire in protest, or roll your eyes in despair and proceed to assign your team to the task?

As an engineering manager myself I concur with all the points.<p>Some of them are difficult though, for example:<p>“optimize the dual objectives of delivering value to the organization and giving individuals problems that build their skillset, impact, satisfaction, …”<p>and<p>“ Managers with excellent execution skills and deep domain knowledge must resist the urge to present solutions to their reports.”<p>To what extent should one allow suboptimal solutions, failure or allow employees to do things inefficiently with their “favorite tech stack” (in cases where it does not add business value) ?

IMHO, we should add “Be Human first, everything else comes later”.

We are putting too much into the word &quot;management&quot;<p>- supervision (did you turn up on time, are you capable of doing the basic functions needed)<p>- coaching (this team needs someone to do X, but I have two Ys. One goes or one changes)<p>- administration (budgets, projects, scheduling etc). Certainly the part most at threat from software - looking at you Project Manahers<p>- strategic decision making (yeah that&#x27;s the part everyone wants to do because it&#x27;s the big bucks with years before anyone can judge you. It is also oddly under threat - not from software but from voting - my guess is most strategy is going to be voted for in twenty years. don&#x27;t like the strategy - binding vote at the AGM for new Board. would save Elon the effort of buying twitter)<p>All four of these (and I am sure MBA text books have better selections) are wildly different, at different levels of an org and need different skill sets and people. just thinking we could ever ask one person to get on with it all is crazy.

Quite naïve. Nothing has been said about what employees need to follow. It is like, everything they can do is perfect, the manager is always the one who needs to work on the relationship, and there are no criteria to select &#x2F; stop working with employees (i.e. those rules will work with any team).

&quot;Facilitate wellbeing&quot;<p>This seems like a reasonable goal, but it is not actionable advice. &quot;Prefer one on one meetings&quot; is actionable, and it tends to enable you, as a manager, to facilitate wellbeing.<p>Forgive me for quoting my book, but this example is worth thinking about:<p>-------<p>Great news! You just raised a round of investment! You are flush with cash! Now you can hit those ambitious goals that you excitedly promised to your investors. In six months, you can go back and tell them that you totally crushed the numbers — not only did you make your milestones, you blew past them. The investors will be so pleased! Now we just need to tell your head of marketing to ramp up the customer acquisition! Luckily she has a genius for this kind of thing. Oh, wait, here she is now, walking over to you, about to say … … oh damn. She just quit? That was unexpected.<p>One-on-one meetings are useful for many things, one of which is that you aren’t likely to be ambushed by a surprise resignation. There are many personal issues that might cause someone to quit, and these will never, ever be mentioned in a group meeting. Burnout? A divorce? A spouse with a job in a different city? A child failing in school? A parent who is dying? Interest in other kinds of work? No one shouts that stuff in a group meeting, but they might tell you when you are one-on-one.

As an introverted high iq nominal eq person with an execution bias I routinely find companies that are keen to move me into senior leadership roles. This post reinforces why I want to work for a good manager but also how I have zero desire to become one.

I’ve been managing a while and a striking difference between profitable established companies versus quasi profitable growth companies is the culture of delivering value. It goes without saying that you want your reports to thrive at work, and as a manager you want to build a great atmosphere and culture.<p>But what I often miss from posts like this is focusing on the value the group of individuals is meant to create (maybe it’s implied).

Many of these principles may work, but they are not the goal.<p>The biggest confusion I see in new and experienced managers is the point of their job -&gt; maximizing their teams productivity.<p>It&#x27;s not clear from the article that the author has realized this.

“Distribute problems, not solutions” makes sense.<p>Making sure that the problems are the right ones to work on is pretty important as well.

The real rules: 1) fuck your team. 2) look good to your superior.

Many of these, could be principles for life.

This is such a great read! 100% agreed!

this seems basic to me

I like how modern society has created a special caste for sociopaths and then secondarily created an entire industry around teaching the sociopaths how to act normal.

This is typical, MBA armchair psychology.<p>Here&#x27;s the real Principals of Engineering Management, according to pretty much every middle manager, director, and vp I&#x27;ve ever worked with:<p>1. Be caviling and pedantic, so you can reinforce your position of petty power.<p>2. Contribute exactly zero code, infrastructure, etc. Basically, anything that actually provides value to the engineers or the customers, you don&#x27;t touch.<p>3. Put at least a couple meetings on the calendar every day of every engineer that they don&#x27;t need to be at to remind them you rule over their time.<p>4. Push nebulous &quot;goals&quot;, then have very set goals when it comes time to promote&#x2F;give a pay raise an engineer and then insist they aren&#x27;t quite there.<p>5. Put engineers on-call instead of hiring support staff, again to remind the engineer that you own their time, even when they&#x27;re not at work.<p>6. Constantly seek updates from engineers, so many in fact that they can&#x27;t complete the work on time, then promptly blame the engineers for not delivering.<p>&quot;Managers&quot; are people with business degrees who failed out of anything technical. Real managers are architects and PMs who actually understand the product and the technical lift it takes to actually make things happen. They don&#x27;t concern themselves with your comings-and-goings, but whether you can get things done or not.