Stuff I Learned at Carta
lethain.com
lethain.com
I had the experience to work in a scale up like Carta couple years ago where the company stoped to hire in NYC/Berlin and as far as I know they shifted their hiring to Philippines.
Fair play, the end of the day the company had their incentive structures to support this decision.
However, after that and other events I just started to do career movements towards companies that I know that I would bring unique features in my position (eg language skills, legal settings, specific regulatory knowledge, local compliance) to be more not entrenched but in a non-constant second thoughts professional relationships in a good sense or be in epistemically different worlds where international competition is irrelevant (eg clearance filters based in nationality, government and military, market that has exotic languages, etc).
I say that because I really do not like of this “Employee as a Service” where line an AWS console you just change the region and spin up labor like some EC2 machine; where in this scenario, you are seen as some expensive spot instance in us-east-1.
Maybe I am being highly defensive, but I do not see hereafter anything I that regard getting better since we have remote work and talent everywhere.
I like this hedging strategy, can probably also apply this to the risk of AI taking over our jobs (licensed professions won’t be going away any time soon).
If you follow the link within the article, he goes on to say:
> The most frequent issue I see is when a literal communicator insists on engaging in the details with a less literal executive. I call the remedy, “extracting the kernel.”
Most engineers I’ve worked with have been “literal communicators.” Of course, both parties can always improve. But part of being a good leader is having excellent communication skills, and that includes anticipating how your audience will receive your message. The bulk of the responsibility is, and should be, on the leader to avoid misunderstandings in the first place.
How do you avoid misunderstandings as an executive when you sometimes literally should hide the information?
I heard many many executives (probably, that's why I am not an executive), a lot of them try to hide information for different reasons. Even the technical one's are trying to keep doors open for interpretation, so that anytime they can change their mind and blame team for the failure, then label them for layoffs
I've worked with two teams where layoffs had to happen. The people weren't happy, but they were at least satisfied that the results were fair and honest. They appreciated my transparency, and worked to train up other members of the group to prepare for their own departure.
If you spend your time building trust and relationships when times are good, and weed out the toxic personalities during those times, then it's better (not easy or good at all) when times are tough. Allowing even the slightest amount of toxicity is completely unacceptable.
If your boss hides information or is intentionally vague to provide an out for themselves, they shouldn't be in a leadership role. They shouldn't be employed at the company.
Being a boss means that 99.999% of your actual job is communicating clearly and openly.
There are "good, successful leaders" but in my experience they are few and far between, and often the "successful" aspect is forced to plateau by the "good" part.
In my humble experience ranging from startups to Big Tech, they are vague with words they use, inspirational in nature, but tries to hide the systematic issues.
For example: I believe our team is now stronger than ever before. How should I interpret it? You were B-player, and hired bunch of C-players, you have just laid off them or did you get rid of bunch of corporate empire builders and now we are indeed in a better position with less bureaucracy?
When I had to lay people off the first time, we had very clear criteria for the layoffs and published them. Anyone who was obviously trying to game the system was immediately dismissed.
Afterwards, I had an all hands meeting to let people know that what we were going through sucks, but that we're in a safer position financially because of it. I then explained all of the financials to keep people from panicking that we were actually broke when we weren't; we had just lost a huge contract to a company that went out of business.
Bad leaders use vapid feel good phrases from Maxwell and Sinek and others like them. Bad leaders use leadership books as a guide to interacting with their employees instead of using the concepts to modify their existing, real personalities. They think those things are a script instead of a series of metaphor.
This can be both true and unhelpful at the same time. “Extracting the kernel” is about putting agency back into your own hands when someone else is less-than-perfect. How do you read beyond the utterance to understand the intent? Will that lead to better outcomes?
Since you sadly cannot force leaders to improve, and sadly cannot usually also pick for yourself perfect leadership, what power do you have to make things better?
And so in that sense, I agree with you - from the perspective of the engineer in this example, yes: try to figure out what they meant and don’t get lost in the details. It’s a good example of not trying to control things that are fundamentally out of your hands.
But the other side is: this blog post (and the linked one explaining the “kernel” idea more deeply) is written from the perspective of the CTO! And it’s framed as a strategy - “encourage your engineers to learn how to intuit what you mean, and not what you say” (paraphrasing, of course).
I think that’s where it rubs me the wrong way. It subtly puts the responsibility for effective communication the receiving end. If we are considering it from a pragmatic standpoint, it’s just far more efficient for the CTO to say what he means from the get-go.
I mean, honestly even with the example: how much harder would it have been for the CTO to say “is it possible to go faster with something off-the-shelf rather than build our own?”
In any case though, you’ve managed to nicely illustrate both of our points, so kudos on that.
That's just laziness on the part of the engineer who says that - it's not only unrealistic, it's intellectually lazy not to understand communication better, and it's going to apply to everyone they talk to, not just executives. It's highly unproductive. It's not serious to say 'I'll communicate how I want, and it's up to other people to understand me'.
And yet, this is fundamentally the position the CTO took.
> That's just laziness on the part of the engineer who says that - it's not only unrealistic, it's intellectually lazy not to understand communication better, and it's going to apply to everyone they talk to, not just executives.
Of course all sides should try to understand better. But one side has a much bigger impact by putting in a little more effort.
It can be true that Lethain is an excellent writer and also a job hopper who doesn’t bear the consequences of his decisions.
https://news.ycombinator.com/item?id=38897363
And I guess related to the section and post on heavily and quickly adopting LLMs: Do you have any thoughts on how to ensure that sensitive customer/shareholder data is not inadvertently mixed in to some new workflow involving third-parties while keeping it accessible for production apps and services?
Keeping confidential/sensitive data from leaking into marketing workflows seems to have been a historical and relatively recent challenge for Carta so would love to hear how you were able to transition from that to securely managing the mentioned level of LLM deployment integrating across the org.
I gotta say, nothing fills me with as much excitement for a job as much as having to have a second job as a Kremlinologist, attempting to scry the motivations of the opaque execs, whose whims come down from On High, either engraved on stone tablets dropped directly into our teams, or brought down to us through three translation layers of middle-management.
I don't like that executives don't have to put in the effort to communicate their concerns and put pressure on the people who already have the pressure of researching, validating, and presenting the solution.
I'm probably going deeper than I should. Still, if the executive asking the question isn't technical, he could direct the question to the executive who's supposed to have a technical background that earned their position in the company. You know, people making decisions should have an understanding of what they are building/selling.
Competent people can often lead themselves in the right direction, especially with the use of copious after-meetings in which everyone tries to interpret the executive’s feedback.
After all, the executives are busy and hard to get access to.
- rely on competent people directly under the executive to guess what that means
- after-meetings in which everyone tries to interpret the executive’s feedback
- the executives are busy and hard to get access to
Imagine having a riddle as feedback in which people from different backgrounds and cultures gather to decipher a meaningful direction.
I think a more generous interpretation of his point is:
1. Everyone (execs and others) needs to translate their true goals and motivations into language, and some people do this better than others. Even great communicators have some delta between what they say and what they truly mean.
2. Thus, it pays to keep this in mind, and think from the perspective of "OK, what is the top-priority, primary thing that this exec is really trying to get across", which can sometimes mean you need to separate the wheat from the chaff of what they're saying.
Frankly, I think this is good advice for anyone on the receiving end of a message. I think this HN guideline deserves a reminder: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith." Thus, you can either interpret his statement in a positive manner about how you can be more effective when on the receiving end of a message, or you can complain about needing to be a Kremlinologist.
Why is the lack of understanding necessarily the exec and not you? It's a victim perspective to say it's all them and you are a victim of them, and therefore have no responsibility for it. It's irrational - the problems of communication are well known (and if you don't understand them, that is also irresponsble, imho). It's you too; worry about you, do your job well, including the work, and joy, of understanding others.
The problem is usually both - both trying to cram the other's words into the framework of their own perspective. Ash and Barry have different perspectives, different agendas and understandings. When Ash hears Barry speak, Ash makes the mistake of trying to cram Barry's ideas into Ash's perspective, where of course they don't quite work. It's like taking parts from one building under construction and trying to use them on another - 'these parts are awful! they don't fit or work at all!'. Of course not.
Always, if there's a problem and you're serious about it, the place to start is with what you can control and are responsible for - it's in the mirror.
I guess our words just don't fit into each others' perspectives.
I’m most interested in the part he doesn’t share: “I’ve also learned quite a bit about venture capital, fund administration, cap tables, non-social network products, operating a multi-business line company, and various operating models. Figuring out how to sanitize those learnings to share the interesting tidbits without leaking internal details is a bit too painful, so I’m omitting them for now“
the hard truth is that most of the time executives are really absolutely clueless and have only shallow understanding of whats going on inside the tech.
and their usual levers are mostly just distributing resources: more resources to this org, fewer resources to that org. and another level is high level program management: checking the milestones, schedules, and timelines. thats it
There it goes. The mask comes off. The sad reality of it is that your average exec views their work, indeed their life, as strictly zero sum. We should not praise this. We should not celebrate this. If anything, we should call this out.
2) 1 person bad does not make all people bad
the people who are in-deep for decades 1) have no time or motivation to write/build a brand^, 2) have political reasons you can't write anything insightful esp while your own house is messy (and it never isn't).
so how to change this?
my approach is to run a conference where practitioners do short ted-talk-like-but-technical talks about their work and learnings while still on the job. but ofc they all have their own motivations to speak.
^ though i will be the first to say that you CAN be primarily self-motivated to write-to-think-out-loud and this is a thing, do not let jaded HN people tell you different
This may go against "SV standards", but I've found that the execs that I most admire at any particular company put in at minimum 4 years - that's the time it takes not just to get a project or two off the ground, but to really instill structural and cultural objectives at any company, and as you put it , to really deal with the consequences of their decisions and refine as necessary. While everyone have some "misfires" in their career, I'm extremely wary of advice from someone that has literally never hit the 4 year mark despite nearly 2 decades of experience.