Ask HN: Is Elon Musk right about firing 85% of Twitter engineers?
A lot has happened since Elon Musk took over and fired around 85% of Twitter's engineering workforces. There has been many controversial news surrounding the whole ordeal.
My question though is STRICTLY about the running and engineering practices of Twitter/X. We have seen Twitter's revenue plummet, and a lot of negative press.
But what we haven't seen is twitter the actual website/platform implode like many engineers predicted.
So I want to ask for the opinions of y'all. Are most companies hiring too many engineers? How were they able to maintain this massive platform despite cutting off so many engineers. Have we over-estimated the importance of engineers? Engineering is generally a smaller org compared to everything else. In your example, has the quality of tweets or whatever it is called now, decreased? Are you seeing lesser engagement? Have your sign-ups decreased? How many new advertisers are you attracting? What about the revenue? Why have all of these take a plunge? As engineers, sometimes, we tend to think of all problems as just software and forget about the large machinery that works behind us to make the product successful. A good analogy is the military. When you think of armed forces, you think of soldiers, the fighters. But, for each fighting person, there is a 5 - 10x the number of support personnel - from logistics, to chefs to camp maintenance to chaplains to everyone in the middle. Similarly, for twitter/x to be successful, it needed marketing, sales, support, content moderation, legal, etc. If you get rid of all of those, then you maybe running a software, but not a business. At the same time, it is true that most large companies run a bit fat. But this is also the first chapter in Mythical man-month, aptly named "The tar pit" and Frederick P. Brooks, Jr does a better job than me in explaining why it is so. [https://web.eecs.umich.edu/~weimerw/2018-481/readings/mythic...] Let's go to https://status.twitter.com to see how it's doing. Oh, right, then. It is absolutely True that most larger tech companies could run much leaner. Almost all projects that are done by many cross functional teams with lots of coordination overhead could be done by a small team of effective senior engineers. Middle management, project/product empires mostly increase overheads with sublinear increase in output. Just maintaining and keeping an already built platform running at its current state doesn't require many engineers. So if that is your only goal then cutting your workforce to a skeleton crew will probably work for quite a while. It's only when you want to add new features, explore new markets and/or find new ways to increase revenue that all those extra people become. I don't know what all those people at Twitter where doing, but I have worked at large companies and many engineers where working on developing possible new products and revenue streams, many of which, for different reasons, never see the light of day. If you're just looking at it from the outside it looks like there are a lot of people literally producing nothing. With all that being said, I'm sure there are lots of companies out there (including pre Musk Twitter) who could lose a bunch of people without it affecting their bottom line or long term productivity. I hear you, but Twitter has been adding new features, no? They introduced community notes, video streaming capabilities etc... If your point was "Twitter has been adding new features despite their engineering cuts", that is incorrect. Community notes and video streaming were all added before the Musk acquisition. The only engineering changes I can think of since the acquisition have been semi-disasters in my opinion: 1. Paying for blue checks. Which for me just means that blue checks are the hallmark of the lowest quality comments and posts now. 2. Renaming from Twitter to X. This one highlights how Twitter is basically unable to even do this correctly without a proper engineering team. Normally you'd implement this so that navigating to twitter.com would redirect to x.com, but Twitter had to do the opposite because changing the primary domain name likely would have touched a ton of underlying assumptions throughout the code, and Twitter was no longer equipped to be able to handle a change of that magnitude. When you build a house you don't keep the builders around once the house is built. You occasionally get a maintenance guy to keep things running but in a well-constructed building the maintenance effort is significantly lower than the initial effort to build it. Why should it be any different for an IT system? Once the system is built, stable and does not require changes (Twitter's feature set hasn't meaningfully changed until recently, ironically post-Musk), it shouldn't require anywhere near the amount of engineers that it needed when it was being developed. Anyone saying otherwise likely has their salary depend on it, which is a common thing around here thus all the noise that was made around the time the firings were announced. I'd say it's more like building a hotel than building a house. Sure you might not need the people that built the hotel, and the hotel probably won't fall down if you just have a minimal maintenance staff. However, if you want the hotel to thrive and grow and be profitable long term, you probably want to hire more than just a couple maintenance guys. Yeah, that's making a lot of similar companies look pretty bad. Especially considering that Twitter's functionality is fairly crisp and works well. I can't say that about say, Facebook and others. Whoever said that Twitter would collapse has egg on their face now. But I wouldn't say we've "over-estimated the importance of engineers" but rather these big tech companies are mismanaged. > But what we haven't seen is twitter the actual website/platform implode like many engineers predicted. Twitter had plenty of issues post-firing. There were outages, features not working correctly, etc. Let's not forget that the firing was a giant f-up, poorly planned, and done at the behest of a mentally ill billionaire. > were outages, features not working correctly, etc. But then it stabilised. While launching new features. Twitter's value destruction has been entirely a function of Musk's personal antics. Not their engineering quality. yup, you've taken words right out of my mouth. I am just curious what was the output of the 85% of engineers that were fired. Mind you, among them were ver senior/director level, so it wasn't just college graduates and middle managers There are still many features not working. I used to search within lists (was like the best feature of Twitter, for my usage) quite a bit -- after Musk gutted the team after taking over, it no longer works, and has been that way for many months. It also had issues before the take over. Firing engineers doesn't have to be a problem as long as leadership understands less work will get done. It's either that or work your team to the bone till they burnout and quit. Both scenarios are pretty typical. He could have just re-assigned them to work on LLMs. But no. Now, after paying many millions in severance, he is hiring an entirely new team to do an LLM model. Twitter is being sued for severance The great thing about this argument for chronic bullshitters like Phony Stark is that it's heads-he-wins, tails-you-lose: * if the platform implodes, clearly the people who were fired didn't build a good platform * if the platform doesn't implode, clearly the people who were fired weren't needed