Headlines about the massive job losses at Twitter have focused on the big numbers — half of the 7,500-strong team left without a position, trying to find employment when their contracts run out on January 3, and in the meantime feeling unwanted at headquarters in San Francisco. (Unless, of course, you’re one of the folks Elon Musk has decided he fired in error, in which case, sorry, please come back, he’s sorry hefired you.)
However, little detail has surfaced about where those cuts came. We know that Twitter’s communications team, which explains to the world the weird whims of its new billionaire owner, has been reduced from nearly 100 to just two. Gizmodo understands that’s now at most one: one of the two staff remaining has since left the company, and it’s uncertain whether the other remains.
But beyond that, it’s been difficult to discern definitively which teams have been affected the most, but analysis of a spreadsheet created by Chandan Maloo, a Twitter staff engineer since May 2021, and filled out by fired Twitter employees touting their skills and roles to would-be employers, does shed some light. The #oneteam Tweeps Talent Directory contains the names of some 830 of those fired in Musk’s 3,750-person bloodbath: that means around one in five of all those cut from Twitter are on the list.
The spreadsheet gives some insight into where the ax fell the hardest. Of the 830 staff let go who have signed up to the spreadsheet, 360 have the word “engineer” in their job title. (A sample of the job titles were cross-checked with associated LinkedIn data to ensure the spreadsheet’s contents were legitimate.) They include 26 machine learning engineers, who may find themselves snapped up by AI companies looking to fill out desks with former big tech staff. The majority of the laid off employees were based in the U.S.
Engineers are the worst affected by the layoffs in sheer numbers, according to the spreadsheet – with the obvious caveat that those filling out their details to be included are self-selecting, and likely reflect the communities in which the spreadsheet is shared on LinkedIn. Engineers are more likely to follow fellow engineers than, say, marketing staff (28 of whom have lost their jobs and filled out their details, per the document), the 14 ex-employees who have “communications” in their job title, or the dozen recruitment staff who have now found themselves out of work. After engineers, product managers appeared to be the next worst affected job title; 24 members of Twitter’s curation team have also added their name to the list.
One Twitter employee who avoided being cut and is still working at the company within the engineering team, says that the proportion of engineering staff included in the layoff spreadsheet “sounds about right” based on his experience of how staffing has changed on the team.
Beyond the material impact on the platform, the halving of the workforce has a human impact, too. Just short of 200 staff members had visa support from Twitter that has now been taken away from them, according to the spreadsheet.
The range of experience among those laid off varied significantly, from some who were in their first year in the tech industry to one senior software engineer with 40 years’ experience, according to the spreadsheet. The LinkedIn profile of the software engineer – who Gizmodo is not identifying to protect his identity – suggests he was at Twitter for less than nine months before he was fired. He did not immediately respond to an interview request.
That has obvious deleterious effects on the stability of Twitter, the website and app, as users are finding out: phantom tweets disappear, follower counts fluctuate, and GIFs glitch into obsolescence. “I would be surprised if those staff were just sitting around all day doing nothing,” says Ian Brown, an information security and privacy researcher and visiting professor at FGV University in Buenos Aires, Brazil. “I would have thought they were there for a good reason, and maintaining a high availability global service like Twitter takes a lot of global resources.”
The layoffs have also massively disrupted the engineering department and its normal workflow. “Our previous software model was very incremental,” says the current Twitter engineer. Engineers would push a service and watch how it performed, he says. They would then find some way to improve it, make those improvements, then push it again. They’d then watch again and find something else to improve. Now, it’s a free-for-all, where major new features are being deployed into Twitter’s code base without prior checking, with the obvious chaotic effects.
It’s an assessment that Foone Turing, a programmer who has previously worked at Google, agrees with. “A lot of known situations can be automated – rebooting machines, cleaning out temp files and logs, and fallover,” she says, “but the problem is the issues you don’t expect.”
Turing says that engineering staffing reductions affect “things like ‘nobody knew that system A was dependent on system B to boot, and system B was dependent on system A to boot, until they both happened to go down at the same time’”. Turing points out that is an example that happened to them at one of their former jobs.
The scale, brutality and way in which the layoffs were enacted makes Twitter’s position even trickier. The way they’ve been enacted could be contrasted with those at Stripe, where CEO Patrick Collison emailed staff to let them know why a 14% reduction in headcount was taking place. Unlike Twitter, Stripe’s CEO appears to have informed his HR department about the imminent layoffs. Staff were also told about the layoffs with a signed email by Collison, rather than an unsigned message from “Twitter”. Musk’s method of laying off staff also compares poorly with Meta’s 11,000-strong layoffs, which CEO Mark Zuckerberg informed them of with an apology while also guaranteeing large amounts of severance to those employees losing their job.