Advice for new principal tech ICs (i.e., notes to myself)
eugeneyan.com
eugeneyan.com
I've always felt staff+ was a bit of a trojan horse on the IC ladder. I understand the role and its importance to be closer to the work. But some companies really seem to love separating you from the work to be done in pursuit of the almighty iMpAcT.
I have seen extremely talented engineers not get their promo to this role because they were "too into coding." And I have advocated for one of them (w/o their knowledge), arguing the project wouldn't exist in it's current form without their technical contributions, and then witness the org retcon their judgment into a staff-level promotion without any significant change in duties.
Makes it hard to take too seriously, honestly. If the purpose of a system is what it does, then it's meant to be a miserly, inertial source of institutional status. One that can get you a lot more money, of course, so play the game as best you can without losing yourself in the process.
Isn't this good for the engineer, though? Presumably they are good at what they do because they like it. So they get a nice promotion and a raise without a change of duties, so everybody wins.
Why promote them out of their skill set, and possibly into a role they will hate and get them fired or willing to quit?
It it usually correlates even more with their ability to portrait themselves as capable.
Thankfully, there is significant overlap between these groups, but sadly not a full circle if visualized as a Venn diagram.
Believing otherwise is psychologically dangerous: if the system is somewhat arbitrary, then that opens up the possibility that their promotion was also somewhat arbitrary and not under their control.
> 25. To get to principal, you need to put yourself on the critical path. To be effective as a principal and go beyond it, you need to actively remove yourself from it.
And
> 26. If you were promoted to principal, it’s because you’ve been acting as a principal for a while
Say you need to keep doing what you’re doing, but also change everything.
Isn't this like a recipe for the Peter Principle?
"The Peter principle is a concept in management developed by Laurence J. Peter which observes that people in a hierarchy tend to rise to "a level of respective incompetence": employees are promoted based on their success in previous jobs until they reach a level at which they are no longer competent, as skills in one job do not necessarily translate to another"However, none of this has much to do with what I said: isn't promoting someone for things they would have to stop doing (as in TFA) a recipe for the Peter principle?
At some point, we have to understand articles and blog posts like the one we're discussing are mostly fluff; an ad for the person writing them. They are self-promotion, there's not much sense in trying to extract valuable lessons from them.
----
I thought of another analogy:
"You're an excellent marksman and a sniper, therefore we promote you to... general of the army!"
But, you could argue, maybe the sniper was already directing strategy, and that's why he got promoted? Nope, TFA is clear about this:
"To get to principal, you need to put yourself on the critical path. To be effective as a principal and go beyond it, you need to actively remove yourself from it"
Just my opinion: one of the key mindset shift that happens before someone steps into staff and principal engineering is understanding that the technical choices you make are tradeoffs. You give up conceits (or one should) that a technical opinion is separated from context, while also getting to deeper technical principles that appear across architectural levels or disciplines (such as queues). You also see how technical systems are inseparable from the people using it, and being a part of it.
When you start seeing that, you start seeing that everywhere. You start to act in a way that is informed from that understanding. You need to be on the critical path in order to do what matters, but you also need to connect those dots that are overlooked.
But that is also one flavor of principal engineering. I tended to do the horizontal influence than deep technical and domain expertise. At early stage startups, people have to wear multiple hats, often exceeding the scope of what they were initially hired for.
Technically that’s true of any step change - a junior developer is expected (loosely) to fix the bugs and do the tasks, to the spec. They should ask for help early and often, and should have enough slack in their estimates from their leads that they have time to mess up to learn. If they continue to do that they will stay junior forever.
They need to learn when to ask, how to estimate and how to avoid certain pitfalls. When they stop asking the basic questions and learn to research, they’ll be promoted to mid level (which they’re already doing). To be an effective mid level though you need to start making mistakes, but in a different way. You can’t be stuck on not being able to install a tool for 3 days, for example.
It’s intellectually dishonest for the author to present collective advice as the their own (“I’ve distilled … My perspective”)
1 - Very few people conduct "proper scholarship", and fail to trace ideas back to their original inception and cite them correctly. This happens time and again in deep learning, where 30+ year old ideas are claimed as "novel" over and over. Many times out of malice by the authors, sometimes out of ignorance.
2 - Peer review in many parts of the industry+research is a joke. Mostly shouldered by early graduate students who don't really know the field well and an incredibly noisy process.
3 - It is common practice now to dump out one's "kitchen sink" of ideas rather than properly refined stuff. Hence the increase in LinkedIn spam, blog spam, arXiv spam style of papers.
Anyway, the article seems very Amazon centric since I have no idea what an L6 or an L7 is. I get that they’re career ladder steps but that’s it.
And having testimonials about yourself on your own website…
The whole website feels like I clicked on an Ad for a person.
L7 is a principal engineer - typically effecting change at org level which will impact many teams
It’s been sad to watch the talent exodus there on my LinkedIn these last 12+ months as these folks flee the ship for elsewhere. So much experience and knowledge just gone and the bar for L7+ with those left has tumbled off a cliff.
I dunno, maybe in Silicon Valley or the US in general, but I think most places outside of that took a much more balanced view of those people even back then, they're just humans after all.
Many people experienced working with those "rockstar" engineers outside of Amazon, "rockstars" who still tried to work as if they were still at Amazon, and the obvious effect of that was that it created a lot of needless friction between the people who saw themselves as "I'm the best engineer because I was L7 at AWS" and the rest of the company.
See recent press on leaked internal doc about folks at AWS whining that they struggle to secure speaking spots for their folks at AI conferences because nobody sees AWS as having recognized leaders in this space.
I think every company that has a "strong culture", no matter what direction, tends to impact people deeply enough that they try to take that with them when they join new places, often creating lots of friction and adding complexity where there shouldn't be any.
>Amazon brain drain finally sent AWS down the spout
You're doing all the politicking and influencing stuff many of us presumably don't like and associate with management roles, while also being expected to be "hands on" and at the top of your technical game. "Nothing is not part of your job", as this article describes it.
Someone who's not doing this, the article argues, "is setting themselves up for failure." Yikes! These are not rookies if they reached Principal IC, but the most experienced team members ever, yet the author still feels the need to say this. Which makes me thing it's a really perilous path.
Seems highly stressful. I'd rather stay a low-level IC. Do we need to move up or out? (In general I mean, I wouldn't want to work at Amazon).
> These are not rookies if they reached Principal IC, but the most experienced team members ever, yet the author still feels the need to say this.
At this level the job is qualitatively different from what went before - you do start as a rookie in this role, and if you only try to keep doing what you’ve done before only better then you’re not setting yourself up for success.
> Do we need to move up or out?
Not to this extent, no. If you are still a Junior after 15 years, that’s a problem and questions will be asked. But if you want to stay in a role where you keep doing what you’ve done before only better, then that’s generally completely fine and the right choice for many people.
Other people here are arguing that you only get promoted to Principal IC if you have been already acting like one in practice. We cannot have it both ways...
And if not, this seems like the Peters Principle. Why inflict this promotion on someone doing well in the other role?
If you show promise but you haven't proven yourself, that's a risky move. You either succeed or you're out, there's no going back to the previous role. I've seen it happen...
Other engineers I've seen (a smaller sunset) have that job description more as an observation of their skills and influence. Their mandate isn't to influence, they just do. They are respected for their vast knowledge, historical success, and insight. So they naturally are heeded by most, and consequently they broadly influence the org.
Both cases sound miserable in their own way, but if I had to choose I'd much rather land in the latter. The latter still involves some politics, but at least it sounds like you're not wasting your life playing stupid games.
I actually don't mind that some people are good at influencing others, through well earned respect, good communication skills and technical chops.
I resent it when it becomes a mandate and some official "badge" in the career ladder. I'm suspicious of these principal/architect types who "parachute" out of nowhere into teams and projects, because it's "their mandate", ask lots of questions, mess with stuff, and then leave and don't take responsibility because "the team owns the project, not them". I've seldom seen this work well. A lot of teams end up politely ignoring what these types say, because they know if you're not a true stakeholder, what you're saying doesn't matter.
It depends on the person, I think. Personally, I often end up doing this sort of work, particularly in smaller companies. I really like it, but appreciate that the vast, vast majority of people I work with would hate it.
For some people, they prefer to lead through influence, rather than through a reporting line. There's a lot of toil in managing people (well), and some really excellent people don't like the core job of management, but are really really strong in some technical area, or have a broad enough perspective and enough personality to convince other people to do stuff.
In some ways, it's the software engineering world's PM, given that you have influence but not direct hierarchical power, and what matters is the amount of teams that you can influence (like sometimes this is through a piece of software, designing and building Airflow was this kind of work).
You worked in the design of Airflow? We're heavy users at my current company.
God no, but I vaguely know a few of the people involved. It was just a good example of a software tool with incredibly large impact.
I'm content being in a senior engineering role where I still get to do technical stuff (and yes, some mentorship as well). Then again, I don't define my life by my career, I'm OK with doing interesting stuff and earning a decent paycheck without climbing up the ladder. Alas! Eventually I will age out of this possibility. I'm already dreading it.
I still write code almost daily however, e.g. for personal projects or in FOSS. For one, I love it as an activity - writing code is stress-relief for me. If I don't do it for some time I really feel deprived. I also think it's necessary to keep the first-hand knowledge of stacks and tools alive and well, e.g. for effectively communicating with engineers in a way both sides will enjoy.
I also do still make a point to e.g. do a MR regularly at work, partly to make sure I know the processes and their pain points, also so I can use my clout to complain if we make engineers waste time with stupid stuff, as processes also tend to accumulate cruft over time.
And obviously I don't think of writing code as a "lowly activity". Quality and skill matter on every level.
For that matter, you can also encounter quite a lot of "the architects don't know how to code"* stereotypes, so the subtle-disdain thing can go both ways (your comment may be exemplary). I try to prove them wrong by having written more and more different stuff than most :-)
* = https://i.programmerhumor.io/2021/11/programmerhumor-io-prog...
Let’s take aligning divs.
What’s the real problem here? The front end looks really bad and it’s causing us to lose potential customers.
Why is that? Maybe we have a shortage of frontend devs. Maybe the frontend code is a mess and people are just hacking in small changes here and there to minimize the time they spend with it. (Maybe both.)
What do you do about it? Fix the alignment now to stop the bleeding. Through that exercise, understand what’s wrong with the front end architecture, get engineering buy-in on an easier approach, and train devs and/or refactor code (efficiently, prioritizing and allocating effort commensurate with the problem), pairing with some devs who have been suffering from the pain and are interested in finally being able to fix it. Advise management on whether we’ll need more frontend talent even after we fix the alignment issue. If so, suggest who might be a good candidate to transition to more frontend work, or else work with sales to lobby for hiring more front end devs even though we have zero headcount budget.
That’s principal level aligning divs.
Should be added to "The Evolution of a Programmer"[0]
[0] https://www.ariel.com.au/jokes/The_Evolution_of_a_Programmer...
Did you miss this?
True role models of mine never were/are L3, L4, L5, etc... They were/are awesome engineers or scientist from the get go. (Like the Johns: The von NEumann and the Carmack.) Experience lets you get better and better, but if you are stuck at a role or organization, the problem might be that you need to find something that you are more passionate about and not necessarily that you are low level because your thinking is not 'independent enough' or you are not 'high-level enough'. At least that is my experience.
I went through a period where I cared a bit too much about an org's leveling system. Not surprisingly, I wasn't promoted, even after giving up some technical work. Eventually, I became really unhappy: I had cut myself off of any work that was fulfilling. It really did feel like I was letting some important part of myself wither away in pursuit of a bullshit hierarchy. I eventually left and joined a startup. I'm slowly remembering how much I like to write programs. Needing to work towards survival is a very tangible goal that has the benefit of making politicking and made-up titles less of a focus...for now.
It's honestly a bit scary how you can lose parts of yourself in these systems. Tread carefully if you aren't 100% into them. Also semi-convinced that some tech middle managers project their own anger/grief over leaving the IC work behind onto those rising through the ranks instead of processing it and finding a sustainable equilibrium.
It has a huge payoff (think 7 figure TC/year for principal+ at big tech), with little personal risk required. It's no wonder people take it so seriously.
Especially if you’re trying to do it as an IC engineer. C suite, directors, and many more probably laugh at their salaries.
Sorry my Omnissiah is acting up again
I’ve jokingly used the example with my own management, “I’d plunge the toilets if they were backed up and became our biggest issue.”
I can’t believe this needs to be said. Who is taking “because I said so” as a (first) reason to make an engineering decision with no justification?
Justifications, needing to say things, etc., are for the weak. The strong get away with shit. That's how you know they're strong!
Or at least that's what >50% of the people I've met throughout my life consider normal. Maddeningly, heartbreakingly, infuriatingly, that number seems even higher among computer programmers
One thing I've found personally rewarding is having to articulate why I believe certain designs / choices are bad ideas, and be ready to propose better alternatives.
If all of my team members were equally experienced, there would be much less need to do this. But it would probably mean atrophying the knowledge underlying my gut instincts, and not being so open to valid alternatives.
If you ask why the vast majority of people have no answer other than “because I like it that way”.
That'd be refreshingly honest.
* Learn new external shit
* Bring that new external shit back to your organization, which requires integration and knowledge transfer
* Inspire people with the new stuff. This means actually talking to people, but its more than that. Its motivating and inspiring people.
That's it. A principle is not a people manager, so there is a lot they don't have to deal with, but they are required to demonstrate and prove their new knowledge in the context of the current organization. This proof can be research papers or MVPs.
Yeah, screw that.
These roles with fancy titles may come with astronomical compensation for an engineer—and rightfully so—but they're essentially buying your soul. You're not an executive, and are still below them on the political and compensation ladder, but you'll easily have 10x more on your plate than an executive. You'll be expected to act as a lap dog for the company for anything tech-related, while you probably will only enjoy 10% of that work. Your guidance will only be appreciated when the stock goes up, while you'll be the first to be held responsible for any technical screw ups.
So, nah. I'd rather continue to enjoy my work, maintain my freedom and peace of mind, and still get paid well enough as a perpetual "senior".
IMO, “nothing is not your job” is odd phrasing that doesn’t really mean “everything is your job,” it’s more like “see something, say something—-in a way that is received constructively and results in positive change, whether through your own actions or others’.”
The unstated corollary is if you’re in a shitty organization that just will not get better, the most positive change you can make for the world is to stop wasting your time helping them and go somewhere better, where you can make a difference.
Be critical, don't be in the critical path, be laid back in an advisory role but be hands-on or you're setting yourself up for failure, work on stuff you enjoy but be ready to justify why it needs a Principal or you're "working on the wrong thing", sponsor, consult, explain to leadership, mentor, code, be present, do not be too present, "feel the pulse", don't attend too many meetings, don't attend too few, gently nudge, don't speak all the time, be careful about staying quiet, etc etc.
Seems like hell. And presumably, you'll get fired if things turn out badly with a project.
Thanks, but no thanks.
“Do not answer a fool according to his folly, lest you also be like him.”
Versus
“Answer a fool according to his folly, lest he be wise in his own eyes.”
The skill is to understand the truth of both statements, and to discern when to apply each one.
Atheist here, so Bible wisdom is especially not useful to me.
Want to go out to dinner with friends? That’s influence without authority right there.
Want to get your PR approved? Influence without authority.
Trying to get your point across to strangers online? Ditto.
That's not really the case.
First of all, software development is—or should be—a collaborative effort. The PRs I create are no more "mine" than the ones I review from my peers. We're all working towards the same goal, and developers shouldn't have to defend or vouch for their work.
Secondly, politics plays a role in every organization, unfortunately IMO. So people who are held in high regard for whatever reason certainly have more authority, and thus influence, to enforce their will over others. Reviews of their code often have a single "LGTM!", or they might even merge without approval.
Similar situations happen outside of software development as well. A highly charismatic person in a friend group has more influence, even though everyone is aiming for the same goal ("get dinner", etc.). An opinion from popular people on tech forums like this one carries more weight than an opinion from someone unknown, even if it's the same opinion. And so on.
So coming back to "principal" ICs in companies, these are mostly political rather than technical roles. The person got to that position because they proved their ability to be influential and lead teams, which generated increased revenue for the company. The company is betting that putting them in a position with more authority, where executives lean on them directly, would lead to even greater revenues.
You’re right about politics, but I think the part where people may vary is the definition of authority. Does being consistently influential make someone an authority? It depends on what that means. They will be believed more often (they have soft power) but they don’t have hard power to command someone to do something. What makes it a gray area is they almost surely have influence with someone else who does have hard power.
Unless the code is a very simple change, the code should at least have the occasional question or suggestion.
Principal ICs sounds like a high stress occupation...
The upside to a more collaborative role like this is you don’t have the stress of having to know everything. Individual developer roles can be more stressful because if you say you’ll solve a problem in a certain amount of time, and then you’re off on your own coding, and things aren’t working out… you’re personally on the hook for the whole thing. Whereas if you’re leading an effort and company priorities shift so people can’t contribute as much, you communicate that to all your stakeholders, and look good for letting higher priority efforts have more resources.
> The upside to a more collaborative role like this is you don’t have the stress of having to know everything.
It's the opposite, actually. The person in these roles has to wear many hats, and have an overview of many areas and teams in the company. The article is explicit about this. They may not be an expert at everything, but they should certainly have working knowledge of each area, have the ability to jump in and steer each ship—whether that involves communicating with each team, removing roadblocks, or writing code themselves—, and be able to communicate all of this in a language useful to executives.
When individual teams are not working well, when multiple teams are not working well together, and ultimately when value is not being produced, it is people in these roles who will be on the hook first.
So the stakes are indeed much higher for this role than for someone working in a single team.
And if you understand the organizational dynamics, being in a position to fix (or recommend the cancellation of) projects that aren’t working gives you more control over your destiny than being just another person trapped in a dysfunctional org.
So I’m not saying higher level roles are easier for everyone, but I am saying lower level roles can actually be harder for someone who has mastered higher level roles.
And in my experience, more senior engineers don’t have a greater risk of being fired for a project going badly because they identify problems to work on that matter and are within their areas of expertise, and evaluate possible risks early and communicate them.
I've seen it backfire spectacularly when a very senior engineer who worked in a critical part of the product was forced into this "because of promotions", then because he was an introvert did it poorly and got a really bad performance review ("underperformer"), got upset and quit. Aftermath: his manager ended up getting fired because of this screwup, but truly that was scapegoating. What's worse is he was happy in his previous role, doing groundbreaking work, didn't want the promotion and wasn't planning on leaving.
I'm not disputing what you say, but in my experience the middle technical roles are the safest. Too junior and you'll be the first to be axed for mediocre performance, too senior and you'll be blamed for failures and fired (sometimes for playing the political game and losing). Meanwhile, the mid/senior programmers doing the work will keep on.
Unless there's another round of layoffs, those upset everything.
It can also be an argument for secret levels. Although I’m not sure how useful that really is in practice.
For someone who does well at influence, it’s not a mandate, it’s permission to spend some time on the nontechnical factors that are necessary to make your work turn out better. And that also means helping others who have good ideas but aren’t comfortable with the influence part themselves.
If that's the case, why is this article needed? Someone promoted to Principal is already savvy, why would they benefit from this advice?
It’s useful to have reference materials to check against, or for things you haven’t worked in recently.
This is like reminding a top doctor "do differential diagnosis". Nope, top doctors already know this, it's redundant advice.
This is like reminding a good thinker they must think about things: they already know this and it's presumably how they got to the position in the first place.
I think this article is a promotional piece for the author's personal brand, thinly veiled as advice for others. It's "look at what I know, and here's why you should pay me". These people love to talk about themselves.
It's technical management. You create impact off the work of others. You tell 10x people to do 10x work and you get 10x the credit when management only is like 1x of talent and effort.
I met a principle engineer who didn't know what a database transaction was and it still fits every description on this site. We shouldn't call these people engineers anymore. They are managers that don't need to do 1 on 1s.
That being said, when I was in the role of a principal engineer at a major tech company - my ultimate responsibility was to get the project to success. What that meant on any given week could have alignment, coding, guiding, or reviewing.
If you think management is like 1x talent and effort, you should try management for a bit.
10x engineers don't need to be principals. The idea behind principals is that they don't do any engineering. They leverage the work of others to become 10x. It's very much considered a "leadership" or management role overall.
A 10x engineer is usually literally a person who can output 10x through his own work alone. As a sidenote, with AI everyone is now 10x. So it's really a 20x thing now.
I didn't understand the point of managers until I got a good one. Perhaps you've just been unlucky
I do know there are entire companies who have not bought into AI yet because the paradigm shift happened so quickly. The AI of 1 year ago could not make someone 10x. The AI of today can and if everyone knew about it, they would be using it.
>I didn't understand the point of managers until I got a good one. Perhaps you've just been unlucky
There is a point. The point is to set direction and to shield the engineer from politics. That is the point. Let me make it utterly clear: The role of leadership and management is REQUIRED. I never said otherwise.
My point, however, is to say that the primary skill involved in setting direction and shielding the engineer from politics and all this "leadership" stuff is this:
it is hard work, but it is not skilled work. The garbage man works hard, but anyone can do his job. There are certain people who do a shit job at being a garbage man and there are people who do a good job, but the overall job is considered unskilled labor. That is essentially what "leadership" is. And like the garbage man, the job is required.
The main difference between the garbage man and the leader is essentially perception. Because leadership is essentially a form of control, perception is controlled. That means leaders are perceived to be better and they are thus paid more but the on the ground truth is that what they do is not skilled work.
The AI of today absolutely does not add 0.5x, I'm using cursor and copilot and they still are usually just a fiddly tool which gets it right half the time. Anything complicated enough to need me to review its work takes longer through series of prompting and correcting its work than if I did it myself, and anything trivial enough for it to one shot its not saving me much time on anyways. All for a costly monthly subscription.
But AI for me is right 90 percent of the time. It’s possible our prompt engineering is different.
> The AI of today absolutely does not add 0.5x, I'm using cursor and copilot and they still are us
You know when people say shit like this I wonder if they ever are able to think from another perspective? Like they say one thing but tons and tons of other people are saying another thing and what gets me absolutely curious is how someone can be so brain fucking dead that they can’t even consider the other perspective. Tons and tons of people say what you say but an equal amount say the opposite.
It would be lovely if all they needed to do was point vaguely in some direction and stop other managers from talking to the team. I could hire much cheaper people for that
Nowhere did I say we needed to point in a general direction and let it go.
As a principal scientist I definitely don't manage people.
Mainly, people managers have power to tell people what to do and when. I have zero power over anyone. No one needs to listen to me. Ever. If people ignore me there's literally no one I can complain to. And no one tells me what to do. If I wasn't competent at my job I'd be sitting alone in an office staring at a wall.
The reason why anyone would ever talk to me is because I have well over a decade of experience more than every other scientist in my org (200+ people) doing very complicated things, at scale, across many areas of AI/ML, and publishing many dozens of papers in top venues. So I know what I'm doing and can find problems, shortcuts, keep them from wasting time, find creative fixes to AI/ML issues, and see connections to business problems. Basically help scientists deliver better features faster. I can find new projects that those scientists are excited about. And I can communicate with product, managers, leaders about very complex technical ideas in ways that more junior scientists cannot.
People in my org come to me because I provide value basically. Not because they have to.
> You tell 10x people to do 10x work and you get 10x the credit when management only is like 1x of talent and effort.
If only! 70% of the work I do is invisible small course corrections here and there across multiple orgs that fix things no one ever hears about but would be disasters down the road. 20% is working on projects that scale with many people. 10% is strategy for new experiments and projects.
For the vast majority of what I do other people get 100% of the credit and my name is a small footnote at best.
Maybe an example would help. Sorry that I need to be a bit vague. A large project recently was going in a particular direction. I saw two months ahead of time that this direction was going to run out of steam for a combination of science and business reasons. It would end up missing metrics. Leadership was asking for something that was broadly correct but subtly wrong. I spent a month finding a new direction that was the opposite of what everyone thought we had to do, finding the right small 1-2 person experiments to derisk the new direction, get senior people to understand it and agree to it, set up the correct relationships to make it viable, and then I had the evidence to convince project leadership and management to shift. A few days later we had a VP review and the response was that they're delighted with the direction, it's far better than they ever though, and this will be a flagship for an org with tens of thousands of employees. That review would have gone horribly without what I did because the problem I saw two months ahead of time would have surfaced and the project would be seen as dying and aimless. But the team got the credit for being on top of things as a whole, not me. As it should be.
> They are managers that don't need to do 1 on 1s.
I have more 1:1s than managers, because they can afford to just tell people what to do. I need to develop junior scientists, keep relationships up, stay on top of other orgs, business priorities, etc.
Managers fail by creating chaos around them. Principals fail by becoming irrelevant.
The skeptics of the role might not have seen how easily and badly a large multiple-team effort can go astray, and the value of someone spotting the problems, and making sure they get addressed, before the product line or company is ruined.
Maybe we should think of the Principal role as complementary? If you're working on a compiler, you're going to be interacting cross-team with hardware engineering team, etc., and spotting lots of things. But someone who is looking at all the teams, and not spending so much time on compiler details specifically, will spot some things the compiler person doesn't. So together you get better coverage of problems that neither alone could spot.
Of course, once we throw differing pay grades into an organization, everything gets more complicated. And people might resent something being called "complementary", if what's bothering them is that the role in question pays better or is considered more prestigious.
Though, the Principal role is in the engineering career track of that compiler writer, if they want that kind of ulcers.
As you go up the chart you have more independence and are less tightly coupled to your manager. By the time you get to principal you should be largely independent. At the same time, you have much more responsibility.
That's just a practical problem. As your manager becomes more senior (director/VP) their scope also increases. They just cannot "manage" you the way someone would manage a more junior IC. Also at the principal level you aren't just bringing value to your manager, but to other parts of the org as well.
In other words, I can't ask my manager "what should I do today?". I cannot even imagine what his reaction would be if I asked that question.
> The manager has to you like as a person and some how believe that all these activities are adding value.
For what it's worth my manager is a great person. But he wouldn't for a moment believe anyone when they say they add value.
It's up to me to find ways to document and express my value. Figuring out how to do this is part of becoming a principal. So I keep notes, I record wins, I make sure that I do things that bring me visibility, that I present new ideas, I contribute to larger roadmaps at the org level, I make sure that other scientists can say good things about me, I help fix problems that other orgs have so that they report I was useful, etc.
In most orgs sadly that's not why people talk to a principal. It's that the process requires it, the manager wanted them to and/or there's you i.e. someone else to take responsibility. Of course they won't tell you that directly.
so i do sympathize with a lot of the negative sentiments about the role here in this thread, and i think that in general there is a lot of navel gazing about the staff+ tech ic roles, there is an actual place for them as tech leads of large projects.
If you have the skill and you're in a larger company they may already have a structure around it you can apply. Otherwise you should talk your manager into trying it, or you should find a new job in an org that does have this kind of role
If you don't have the skill, you should get the skill first, by making your team more effective, then make your team and all related teams more effective together
Then your case is the exception, not the rule. To reach the level of principal, you generally have to be recognized for delivering high impact. Visibility is not just ego, it is how organizations perceive value. If your work earns you no visible credit, then no one really knows what you contribute. And if no one knows, how could anyone justify promoting you in the first place.
That is the ideal. In the ideal world, principals are elevated because they have a visible history of making the system better. They build frameworks that others rely on. They turn chaos into structure. They guide teams through impossible projects. Their reputation is not something they chase, it forms naturally from the wake of their work. In the ideal, visibility is the residue of real impact. People talk about them because their fingerprints are on every success.
But the corporate world rarely functions on ideals. In the real world, power accrues to whoever is closest to power. Titles often flow through social gravity more than technical merit. Some people climb because they deliver, others because they simply survive long enough to become unmovable. The higher you go, the more politics matters and the less evidence is required. Impact becomes subjective. Influence becomes reputation. And reputation, once earned, decays slowly.
In that reality, being invisible is not a liability. It can be a strategy. A principal who keeps their head down, avoids controversy, and stays on friendly terms with the right directors can outlast a dozen brilliant but abrasive engineers. The irrelevant survive because they are not a threat to anyone’s ego. The company quietly carries them, paying tribute to their title while forgetting their function.
Even the ideal, though, cannot escape the need for visibility. A principal who does great work in secret still fails the fundamental requirement of leadership: to be seen. Influence requires perception. You cannot guide a culture if nobody knows you are there. Quiet impact might keep systems healthy, but it does not create belief, and belief is what organizations promote. The best engineers learn to make their results legible. They translate their impact into stories others can tell. Without that, the work disappears into the background noise of everyone else’s effort.
So there are really two systems running in parallel. The first is the ideal, where promotion is earned through visible excellence and quiet authority. It demands both impact and awareness. The second is the reality, where promotion is often granted through time served, connections maintained, and an ability to avoid friction. The ideal rewards contribution; the reality rewards endurance.
You can succeed in either system, but they ask for different currencies. The ideal asks for mastery, courage, and the discipline to lead by example. The reality asks for patience, diplomacy, and the instinct to stay useful enough but never threatening. One builds respect. The other builds stability.
And most companies, if we are honest, prefer stability. Stability requires engineers to act the way you describe but talk the way I do. You talk about the ideal, you even believe you walk it, but because no one can see your impact, no one can tell the difference.
I've held the ranking of staff in many companies. I've interacted with principals, staff, and distinguished engineers and I can tell you visibility is required to fulfill the ideal. If visibility wasn't there, than the person earned the rank through other un-ideal means.
Of course you need visibility and impact. But it's not the kind of crass taking credit for what other people did that the original poster talked about.
As a principal you need to actively build visibility because so much of your work is normally invisible. It means being an active participant with discussions with leadership, clearly being the person that sets the agenda on something, becoming someone who other principals turn to on a topic and then report to their managers, leading the conversations on a topic, being the person that people can bring a hard problem to and then see results, picking up on a business need first and finding a way to deliver on it, etc.
In the example I gave for the quiet change in direction of that project, there are many ways in which I get visibility. I told my manager this was a risk and I have a strange solution. I told the manager of that project. The other principals that I needed to involve and their directors know I pushed this. My VP knows because no one talked about that kind of feature until I did. I checked how crazy this direction change would be with more senior scientists and product people. I actively make sure that these gentle communications happen, and in a sense they're natural, because I'm taking the lead on something.
But no one is going out of their way to say "I take credit for all of this work that everyone else did".
> A principal who keeps their head down, avoids controversy, and stays on friendly terms with the right directors can outlast a dozen brilliant but abrasive engineers.
I don't think that abrasive people should become principals until they change their ways. There's so much cross-org coordination that you need to do, if you're abrasive, that's going to hurt everyone.
That doesn't mean you should be a pushover. I don't back down from technical arguments if I know I'm right and have the data to back it up. I have gotten into deep weeks-long disagreements reported far up to chain. But it's important that you can still go have lunch with your peers and collaborate on other topics even while you're trying to show that they're totally wrong in one area. That's part of building trust.
> In that reality, being invisible is not a liability. It can be a strategy.
A strategy to be fired. This is not viable.
There is no tension between "quiet authority" and visibility. If you're invisible no one will ever come to you. Visibility is what consistently demonstrates to people that talking to you will make their lives better.
Not true at all. Plenty of invisible people in the world who don't get fired. You're ignorant.
Your initial post talked about lack of visibility, now you're talking about it as if it's required comically to the point of claiming that "invisible" people are absolutely on the path to get fired or that people who keep their head low and don't grab attention isn't a strategy. Ever heard of the saying the tallest blade of grass is the first to be cut? Also have you seen how Xi, the current leader of China rose to power?
Part of being a principal is the perception of principled and expert reasoning/expertise. That means being able to flip what you're saying on the fly so that other people continue to perceive you as an expert who knows what they are talking about. Did you say something that was completely off and wrong and it made no sense? How do you hide this from the people above you?
You expertly did this with your response. First you say you take none of the credit, now you say taking credit and being visible is required. Expert pivot and THIS is truly the primary skill of a principal engineer and you display it unequivocally.
You don't even need to actually be responsible for all the changes you claimed to have influenced. You just need people to perceive the reality as if you were responsible. And I see this kind of BS in many, many engineering organizations.
A lot of this is self delusion too. These high level strategies and directions aren't hard to come up with. Likely tons of lower engineers have thought of the method too, they just don't have the "visibility" or the time to actually drive that direction into the organization. Some people tend to think the ideas they have are genius and that these ideas are what makes them "principle". No. It's mostly politics that puts you into a position to take credit for ideas anyone can come up with.