Cruel Changes at Twitter
👋 Hi, this is Gergely with a bonus, free issue of the Pragmatic Engineer Newsletter. We cover one out of five topics in today’s subscriber-only The Scoop issue. To get this newsletter every week, subscribe here.
Last Thursday, I covered the turmoil at Twitter, of how people worked long hours through the weekend and how most expected layoffs of about 50%.
On Friday, 4 November, Twitter started to let go of about half of its staff, around 3,700 people. It’s extremely rare to see layoffs of this size; some of the largest layoffs I’ve covered this year were no larger than about 30%. It’s even rarer for such deep cuts to be executed so quickly. Well-executed layoffs usually take weeks to plan, and I’ve yet to see such deep cuts put together in such a short timeframe.
Tech journalists Casey Newton and Zoë Schiffer wrote the article, Twitter, cut in half, on how the layoffs happened. I recommend reading the full piece.
At 8pm PST on Friday, employees were told by email that they’d be notified by 9am on Saturday as to whether they were laid off or still in a job. Elon Musk did not communicate with employees during this time, and would also not sign the final email letting people go. As Platformer wrote:
“After the layoffs, we asked some of the employees who had been cut what they made of the process. They told us that they had been struck by the cruelty: of ordering people to work around the clock for a week, never speaking to them, then firing them in the middle of the night, no matter what it might mean for an employee’s pregnancy or work visa or basic emotional state.
More than anything they were struck by the fact that the world’s richest man, who seems to revel in attention on the platform they had made for him, had not once deigned to speak to them.”
Who got laid off seemed to make little sense. Some software engineers who worked over the weekend to ship the Verified functionality on a tight deadline were let go. Another software engineer refused to do any work during the weekend and was expected to be fired, but they were not.
Seniority seemed to have played a role in who stayed. According to my sources, in Europe, only senior-and- above engineers were spared. In the US, most staff-and-above engineers were also not let go, as per my sources.
It took only one day for Twitter to start calling back some people who were laid off, but whose expertise was still needed. Tech reporter Casey Newton broke the news that Twitter started calling back people on Sunday, 6 November, barely 24 hours after dismissing them. As per his reporting, managers were told to nominate people who could help if they returned. iOS and Android engineers were high on the list.
On Monday, 7 November I talked with a software engineer who received such a call back that same day. They said they got a call from an unrecognized number Monday morning, inviting them back with a deadline of less than 10 minutes to decide. This person declined the offer.
Talking with more employees, I was told the calls became more frequent, and that most software engineers and engineering managers were saying “no” to this offer with a deadline attached. Employees warned each other to not answer calls from unknown numbers. People also looked into whether Twitter could force them back to work as it just fired them and they were technically serving their notice. Employees came to the conclusion that even if they accepted, they would be entitled to the severance the company had already communicated that it would pay.
I find it staggering these layoffs were so badly organized that almost immediately some of the people affected had to be called back. This suggests those who put the lists together of who to fire, failed to consult those on the ground who keep Twitter’s services running.
The first few days after layoffs were predictably quite messy. The day after the layoffs, engineering managers still did not know who on their teams was staying or going. Some people turned to query LDAP to see if the SSH keys for those usernames were still active. If not, that person was gone. If yes, they were still around.
I talked with a software engineer who was assigned to be oncall for systems they did not know about. When they tried to get more information, they found the team which built the service had been let go.
I talked with another software engineer based in Europe whose manager was let go. Their new manager, now based in San Francisco, organized a group meet and greet meeting for 1am in Europe, with more than half of their new group based in this region. Barely anyone from Europe showed up.
Predictably, resignations kept following all week. After any layoff, it’s to be expected that more people leave. In my experience, a rule of thumb is to expect about half of the amount of people let go to leave voluntarily in the months after a layoff. For example, when laying off 10%, it’s sensible to expect another 5% leave afterwards. When letting go of 50% of staff, it’s not unreasonable to expect another 25% to leave.
So, the resignations kept flowing in. A good portion of principal engineers either handed in their notice or are in the process of doing so. Many of these people had been with Twitter for 5+ years. Some built the first versions of the systems Twitter runs on.
I am told that managers grew increasingly worried about the stream of resignations and how few people accepted offers to come back to the company.
On the night of Wednesday, 9 November, the Chief Information and Security Officer, Chief Privacy Officer and Chief Compliance Officer all resigned.
Experienced software engineers leaving should be worrying because of the complexity of Twitter. Talking with long-time Twitter engineers, they told me the site has a complex architecture for a very good reason. It was built to allow for safe and fast tweaking of any part of Twitter, making it easy for developers to iterate.
However, Twitter was not built to make architectural changes easy, and structural changes are admittedly difficult and so need to be done with caution. Twitter has on-premise data centers and I’m told that deploying Kubernetes on top of this was already a massive undertaking thanks to all the custom infrastructure.
Twitter is unique in how configurable the infra team has made infra-level tweaks. There are multi-level feature flags used at the infra layer, which is highly unusual as feature flags usually live within the application layer.
Elon Musk announced the launch of features already built, for which the whole team was fired. On Twitter, Elon Musk went on an announcement spree, sharing how Twitter will launch longform messages and better search. I’m told this came after he saw internal demos of these features, which were almost ready to launch.
The longform messages feature is called Notes, internally, and has been in the user testing phase. On Monday, 7 November, a director asked about how feasible it would be to ship this feature to all users. The engineer who replied said that it’s hard to tell, given the team that built all of it – the backend engineer, iOS and Android engineers and engineering manager – have either been fired or quit. Again, to me it shows just how chaotically the layoffs were orchestrated.
Most remaining engineering teams outside of the San Francisco headquarters have one thing on their roadmap for the next year: cost cutting. Out of the three data centers operating, the engineering team is investigating shutting down one, to leave the company with two DCs. There’s also a push to significantly reduce their spend on Google Cloud.
Going forward, managers are expected to have 20+ person teams and spend at least 20% of their time coding. One of the very surprising changes for managers and engineers is how the management structure is changing.
Most L6 engineering managers who were not let go, and were line managers, were instructed that they are to be individual contributors. Managers now have at least 20 reports, and are also expected to code 20% of the time.
The above changes are like nothing I’ve seen at any tech company. Most don’t have managers take on more than 12 reports because it’s hard enough to give all of them the attention required. When I was an engineering manager at Uber, I purposefully did not code and I felt close to burning out when I was managing 14-16 people directly.
I expect most managers to do a poor job of managing people, do some alibi coding on unimportant tasks to make Elon happy, and burn out during the process. And better managers know this, already.
One new engineering manager was honest with their direct reports about this change. They told reports that going forward, they cannot manage their career, they cannot take care of feelings, and they’ll have very little time for them, as they now need to be coding.
Twitter is getting rid of remote work, instructing employees to immediately return to the office. Elon Musk sent out his first email to all staff at Twitter just before midnight PST, on Wednesday. The email has the subject “Difficult times ahead”, painted a dire picture of Twitter’s economic prospects and said that Twitter will not survive the upcoming economic downturn without subscription revenue becoming roughly half the company’s revenue.
The email closed by saying that starting on Thursday 10 November, everyone would be required to be in the office for 40 hours per week. The only exceptions are people who are physically unable to travel to an office, and exceptions that Musk personally approves.
This message caused immediate uproar among employees. This is because not only did the company announce permanent Work from Home in 2020, but hired a large number of employees as full-remote employees as per their employment contract.
I am told the HR team has been caught off guard by Musk’s email and are scrambling to put a formal policy and communications in place for this change, as there is currently none.
Internally, on Slack, a current legal employee wrote that their personal view is that Twitter employees do not have an obligation to return to the office, especially with no notice given. This employee also reminded peers that Twitter’s ex-security lead, ‘Mudge’ became a whistleblower by reaching out to Whistle Blower Aid.
In the middle of the chaos, the sweeping changes are creating opportunities for people who are ready and willing to step up. While many are leaving and just as many are looking for jobs, there are some who, amid this chaos, are leaning in. A Director of Product shared a picture of herself sleeping in the office and has since become a key figure in coordinating with Elon about new features, and making product announcements. A staff engineer has also been promoted to manage a large group, without having been a manager before - likely a testament on how much Elon values software engineering experience over management experience.
What is happening at Twitter? My head is spinning from hearing all of the rapid, cruel, and likely regulation-breaking changes at the company. What is happening?
I think back to Musk saying that he wants to reduce headcount by 75%, something he later walked back on. But it sure seems he is hell-bent on getting the workforce reduced by this amount. After letting go half of the staff, he’s putting in place changes that are designed to create voluntary attrition.
My view is Musk is playing with fire by cutting down on headcount so quickly, then encouraging more people to leave. We could see Twitter experience larger outages in the coming weeks and months, stemming from software engineers making changes to systems they don’t understand and not having colleagues with expertise to turn to, as they have left.
Also, while he is reducing headcount, he is putting the same culture in place as at his other companies, like Tesla or SpaceX. At these places, managers have 20+ directs, managers are expected to be hands-on, and working from the office is mandatory.
While Musk has all the right to implement processes and management structure as he sees fit, I find no excuse on how he is doing all of this. He offers no compassion or respect for employees who helped build Twitter, and employees feel that he is outright hostile towards them. Just the fact that in two weeks, he managed to send a single email to employees - and that email was an ultimatum on returning to the office effective immediately - summarizes his attitude towards employees of the company he bought. Elon has broken all unwritten - and may written - rules of how tech companies operate.
Tesla and SpaceX are both companies where hardware is arguably just as important - if not more important - part of their success than software. Elon Musk turned both companies into roaring successes with a similar management style which he’s bringing to Twitter. It will be educational to see how this same management style will work at Twitter, which is a pure software company, which does not have an inspiring mission or challenging problem space that could compare with Tesla or SpaceX.
This was one out of the five topics covered in this week’s The Scoop. A lot of what I share in The Scoop is exclusive to this publication, meaning it’s not been covered in any other media outlet before and you’re the first to read about it.
The full The Scoop edition additionally covers:
- A change to The Scoop. A bid to bring you more positivity and interesting news, to counterbalance bad news about layoffs.
- A Staff+ peer group success story at MongoDB. A staff engineer noticed they rarely worked with similarly experienced colleagues. So they started a peer group and have shared how it’s going.
- Plain: behind the scenes of an API-first customer support platform. The two-years-old startup just raised $6M and I talked with their CEO and engineering team to get some more insider details. What tech stack do they use? How easy or challenging was the fundraising? Exclusive.
- Success stories. A senior software engineer bouncing back from burnout, a healthcare medical assistant turned software developer, and an eye-catching parental leave policy.
Update on 12 Nov: the article previously stated that no staff and above engineers were let go. Updated to ‘most staff and above engineers were not let go’ after getting confirmation of one such layoff.
Featured Pragmatic Engineer Jobs
- Full-Stack Engineer at Farmlend. £85-95K + equity. London.
- Senior Backend Engineer at Farmlend. £85-95K + equity. London.
- Senior Full Stack Engineer at Perfect Venue. $150-180K + equity. San Francsico or Remote (US).
- Senior Software Engineer at AI Build. London or Remote (Europe).
- Senior DevOps Engineerr at AI Build. Remote (US).
- Full-Stack Engineer at Vital. $70-120K + equity. Remote (Global, within 5 hours of GMT).
- Backend Engineer at Vital. $70-120K + equity. Remote (Global, within 5 hours of GMT).
- Technical Lead - Platform at Vannevar Labs. Remote (US).
- Senior Software Engineer, Fullstack at Vannevar Labs. Remote (US).
- Computational Geometry Engineer at AI Build. London or Remote (Europe).
- Senior QA Engineer at AI Build. London or Remote (Europe).
- Lead Backend Developer at Cineville. €53-79K + equity. Amsterdam.
- Senior Software Engineer, Distributed Systems at Mixpanel. $200-270K + equity. New York, San Franciso, Seattle or Remote (US).
- Senior Software Engineer, Fullstack at Mixpanel. $200-270K + equity. New York, San Franciso, Seattle or Remote (US).
- Principal Engineer at Shoplift. $185-205K. New York.
- Senior Engineer at Sixfold AI. New York.
The above jobs score at least 10/12 on The Pragmatic Engineer Test. Browse more senior engineer and engineering leadership roles with great engineering cultures, or add your own on The Pragmatic Engineer Job board and apply to join The Pragmatic Engineer Talent Collective.
Want to get interesting opportunities from vetted tech companies? Sign up to The Pragmatic Engineer Talent Collective and get sent great opportunities - similar to the ones below without any obligation. You can be public or anonymous, and I’ll be curating the list of companies and people.
Are you hiring senior+ engineers or engineering managers? Apply to join The Pragmatic Engineer Talent Collective to contact world-class senior and above engineers and engineering managers/directors. Get vetted drops twice a month, from software engineers - full-stack, backend, mobile, frontend, data, ML - and managers currently working at Big Tech, high-growth startups, and places with strong engineering cultures. Apply here.