I worked at SpaceX. It was the most efficient company I've ever worked for | Hacker News Reader

I worked at SpaceX. It was the most efficient company I've ever worked for

businessinsider.com

38 pointsamichail32 comments

Open articleView on HN

His prior experience consists of two government focused jobs and self-employment.

Of course SpaceX is the most efficient company he’s ever worked for.

Would it technically also be the least efficient company he has worked for?

He worked for two other companies.

But they (perhaps) were a low bar.

Well, one was the FAA, which isn't a company in the conventional sense. Freddie Mac is marginal (it's strictly speaking a company, but basically a state enterprise). This is a bit like someone who had previously only watched television writing a review of the first book they read; they just don't really have the experience to usefully discuss it.

I also worked at SpaceX - they are not efficient. They refuse to buy software or use cloud solutions, their existing software systems have immense bloat, and their infrastructure is held together by 1 overworked IT guy, and if he left half the factories at the company would shut down overnight.

That sounds pretty damn efficient, although risky.

The real-world outcomes from what I described is that it takes teams of dozens of software developers being hired to either reinvent the wheel or spend all of their time managing tech debt rather than solving actual problems.

Serious question, what is costlier, for a critical high-risk long-gestation operation like space travel - paying 1 IT guy to stick around, or having to go through an entire transformation/ migration/ contract a custom build when your third party cloud intrastructure inevitably goes under, or deprecates, or fails to meet your needs, etc? Granted, there is probably a better third way

depends on the bus factor...

The former is costlier. It's litteraly a gamble that your one IT guy won't:

- just leave

- get sick (including burn out)

- move to another country

- get hit by a bus

It just doesn't make sense in the long term.

PS: bot even counting that the negociating power of that single guy is incredible.

I used to worry about "bus factor", but after seeing countless "indespensible" people leave abruptly and the companies recover just fine, I don't really think it's an issue. It generally sets objectives back a few weeks and that's about it.

I knew an entire custom manufacturing company with 40 employees run with one main engineer, who worked 50-60 hours to complete his work so the employees in the shop could run the machines and do assembly.

He died one day suddenly at 50 years old. He had been the sole engineer for 10 years.

A week later I was training his replacement on how to use the PM tool I provide for his company. Same office, same chair.

It then comes down to whether you can afford to be off for weeks.

Another aspect of this is the hiring of a new guy: I've seen it times and times, where the company doesn't have anyone to evaluate candidates, so they end up peddling more time and money to even get to the point they can start interviewing.

Of course it all comes down to the balance between how much you save keeping a single guy on staff, vs how much it costs when shit hits the fan.

I gave the danger of the bus number one lecture to a bunch of managers once, and had one pushback pretty hard by saying that what he was working on was so complex that no one else could possibly learn it in time to be useful for the next release (every release...).

They bought me lunch a few months later to apologize after having the unique person die and leave them with no backup.

I have a relative who just retired and was replaced by five people who started a year ago. It is extraordinarily difficult to replace decades of knowledge and wisdom.

My buddy works at Tesla here at the gigafactory in Berlin and says the same thing. It’s all on-premise run using open stuff. Which isn’t horrible but the culture seems to be we will build it ourselves.

I love the idea of it being culturally okay to walk out on any meeting that isn’t adding value in the middle of the meeting. If that was culturally acceptable in my role, I would go to a fourth of the meetings I currently attend.

Master the art of sending "getting pulled into another call - have to drop" chats

I would hate the idea that my primary purpose is to get pulled into random meetings:

> Instead of having a manager and a list of tasks, I was asked to sit in on meetings and add value to projects aligned with my interests or expertise.

Disorganised meeting structure with people coming and going at random? How could you possibly get anything done, as a contributor?

I feel like this is what working from home has enabled in some places. Much easier to gracefully leave an online meeting.

This person has worked for exactly three companies. Why would I care about the most efficient company he's ever worked for?

This person was anxious and directionless for his first six months on the job. Rather than seeing this as wasteful and cruel on the part of the company, he thinks it is praiseworthy.

This person has seen entire rooms walk out of meetings with the customer. Rather than seeing this as insulting and capricious, he thinks it is refreshingly candid.

This person thinks SpaceX is "probably the most accepting work culture [he's] ever participated in." Yet SpaceX is plagued by accusations of sexual harassment, bullying, and racial discrimination. It is currently being sued by the DoJ for discriminatory hiring practices.

This person claims to be in awe of how well SpaceX is run, yet he left after barely more than two years "because [he] felt [he] was no longer learning." Hardly the actions of an eager pupil.

This person observed that most people at SpaceX don't even last two years, yet doesn't seem to recognize that as a problem.

It's no mystery how this bizarre puff piece got written:

>From there, I started my new project, Inheritance Art, in early 2022. We're working on various projects — from crypto to large language model services and even creating our own AI models. I'm enjoying the challenge, but I still look back fondly on my time at SpaceX.

Dude's hawking NFTs. His pitch is "I learned from the very best and will run my project just like Daddy Elon taught me."

Well summarized.

> I thought I would get fired every day during the first six months. There was no one there to tell you what to do. In the military, there was a change [sic] of command. At SpaceX, you ask yourself, "How do we do this thing? How do we do it safely? How do we do it so that it's a good value for our customers?" Then you figure it out. [emphasis added]

This guy was never a full-on member of the active duty military or he'd have a different take on this. The cadet life is very different from active duty. As a cadet, much of your life is already detailed for you, even down to when to get up, when to eat, when to exercise, when to relax (this changes over the years as a cadet, but it's still very regulated). If he'd been able to serve in the active duty, having graduated from West Point he would have been an officer.

Officers do not just look around and wait to be told what to do or ask their CO to hold their hand and walk them through their day. They have to figure it out, too.

Hell, even as a cadet you don't get coddled and still have to figure things out.

And the more senior you get, the less and less of any kind of guidance/handholding there is. Because in wartime, you're going to get your commander's intent, some guidance on your objectives, and then be expected to manage the ensuing chaos as the other guy tries his level best to make you fail and/or die.

"There was no one there to tell you what to do. In the military, there was a change of command." With all due respect - in close to 30 years being a contractor and now executive, I've essentially never had anyone telling me what to do. When you're decent at your job - this is how people roll.

SpaceX is a defense contractor and Mars is the cover story.

Just read up on Elon's cofounder and their work on SDI.

https://wikipedia.org/wiki/Michael_D._Griffin

https://wikipedia.org/wiki/SpaceX_Starshield

Musk has no intention of colonizing Mars - or if he does, he's dumber than I thought. The Musk secret recipe has always been about suckling on federal contracts and subsidies to get rich. Just imagine the payday on the exorbitant international contracts necessary to send 3 people to Mars...

marketing

This is a conversation with a /former/ employee

The author is doing marketing their own online reputation.

This feels like a LinkedIn post where someone toots their own horn by talking about how much they learned at awesome previous employers. It’s a LinkedIn Post that escaped to Business Insider.

i am unable understand how do they figure out what to work on ?

How does your manager figure you what you are supposed to work on? What information or ability do they have that you don't? Why can't you do what they do. Ultimately there is only one thing that is missing...responsibility. You aren't responsible. Your manager is. But there isn't any reason you can't be.

Spacex, particularly in the development of Starship, is an example of the inappropriate use of an informal iterative project management and design approach. In other words it's a case of Agile, which is often the best kind of project management for software, infecting places where it shouldn't ought to go.

In the end when you count up 33 rocket engines per iteration, it's going to end up being a lot more expensive than what more conventional projects like Blue Origin will cost.

Reply on news.ycombinator.com