In May, Syndio eliminated a number of roles in a company redesign, part of a shift from a software consultancy toward what its CEO calls an AI-centric technology company. One of the people let go was a labor economist who had been there five years. Syndio's CEO, Maria Colacurcio, expected the colleagues she kept to be glad of bigger roles in a tight job market. Within a few months, several of them resigned. In August, he applied for one of the newly opened roles and was rehired. Coming back, he told her, felt like "sneaking back in the house after I got kicked out."1

Colacurcio wrote the story up herself in Fortune, which is rarer than it should be. "We moved really fast," she told Business Insider. "The technology wasn't quite there yet."2 In her own essay she does not call the redesign a mistake. The company had to change, she says. What she learned was narrower: the danger of chasing a new business model "before fully understanding whether the skills, experience, and institutional knowledge you already have can evolve with you."1

The verdict arrives in pieces

Look at how the evidence reached her. None of it came as a review of the decision. It came as resignations from people she had counted on keeping, and then as an application from someone she had let go, for a role in the structure that replaced his. Each of those events goes to its own desk, in recruiting or in HR, and gets handled as what it is on the surface: an exit, a hire. Colacurcio chose to read them together and to write down what they said about May. That connection is easy not to make.

Amazon is the larger case. It has cut more than 30,000 jobs across multiple rounds of layoffs over the past year, and according to recruiter emails reviewed by Business Insider, it has been contacting eligible former employees, some of them laid off, about open roles. Recruiters working with an organization led by AWS vice president Swami Sivasubramanian reached out under what one email called "Swami's Boomerang Reengagement Initiative," aimed at former AI and machine learning staff.3 A spokesperson, Haley Silva, said the company is interested in former Amazonians "for open roles" and that "rehiring former employees is a normal, long-standing part of how we recruit."2 That may be entirely accurate. It also means the rehire of a laid-off engineer gets filed as ordinary recruiting, and whatever it could say about the cut goes unread.

A layoff is graded on the day it is announced. The evidence about whether it was right shows up months later, one resignation and one rehire at a time, and nobody is assigned to read it.

How common is it?

The staffing firm Robert Half surveyed U.S. hiring managers this year and reports that 32 percent of those who eliminated positions after implementing AI have since added them back.4 Robert Half sells staffing, so it has an interest in that finding, and its public summary does not give the question wording. The broader "boomerang" figures in circulation cannot settle the question either way. ADP Research puts returning employees at 35 percent of new hires in March 2025, and Revelio Labs puts returners at 3.4 percent of people starting new positions at large companies in 2023, a tenfold gap that appears to come largely from different definitions and different employers.56 Neither dataset separates people who were laid off from people who quit.

Before the next round

None of this argues against ever reducing headcount, and Colacurcio does not argue it either. It argues for deciding in advance what would count as evidence that a cut was wrong, and who will look for it. Whether AI was the honest reason for a cut is a separate question, taken up earlier this month. This one is about whether the cut itself held up.

Before the next reduction, write down what you expect to be true twelve months out if you cut the right things: who you expect to stay, what work you expect to go away, what you expect not to have to buy back. Then name the person who will check. If nobody owns that check, the only verdict you will ever get is the one that walks back in through recruiting.