ClickUp Cut 22% and Called It AI. The Ex-ClickUp Pool Is 221.

ClickUp cut roughly 285 to 290 people, about 22% of its ~1,300-person org, and CEO Zeb Evans went on X to insist it was not cost-cutting and not weak demand. If you run engineering hiring at Linear, Notion, Height, Asana, Monday, Airtable, Coda, or Smartsheet, the memo tells you exactly which sub-pool just hit the market, and which one Evans is paying $1M salary bands to keep.

The window to intercept the right half of the diaspora is this week.

What actually happened at ClickUp

ClickUp cut about 22% of its ~1,300-person workforce on May 21, 2026 and Evans framed it as AI-agent substitution, not a cost move, redirecting savings into million-dollar salary bands for survivors. This is the third round in four years (7% in 2022, 10% in 2023 at ~900 headcount, then this one), which changes the average tenure of the people who just left.

The key facts, from public reporting:

• ~290 people cut, 22% of ~1,300 headcount, announced May 21, 2026 by Evans on his personal X account.
• ClickUp disclosed roughly 3,000 internal AI agents deployed across departments, a 3:1 agent-to-employee ratio.
• Evans, on X: "This wasn't about cutting costs. Most savings from this change will flow directly back into the people who stay. We'll be introducing million-dollar salary bands."
• The memo split the future workforce into three buckets: "builders," "system managers," and "front-liners."
• ClickUp was co-founded in 2017 by Zeb Evans and Alex Yurkowski and last raised a $400M Series C from Andreessen Horowitz and Tiger Global at a $4B post-money valuation.

The InformationWeek "early August 2026" framing appears to echo this May 2026 round rather than a distinct second cut. Either way, the pool on the market right now is the same pool, and it is bigger, more senior, and more scattered than the coverage suggests.

3,000

Read the denial, not the headline

Evans denying that the cuts were about cost or weak demand is the sourcing tell: the pool leaving is disproportionately the engineers he does not consider "AI-fluent," not the staff+ ICs he wanted to keep. Recruiters chasing "ex-ClickUp AI/agent engineers" will fish an empty pond. The real pool is real-time collab, workflow, permissions, and integrations engineers, plus the CX-adjacent tooling teams Evans coded as "front-liners."

Three signals reinforce this reading:

1. Evans' own bucketing. "Builders" and "system managers" oversee AI systems and automate their own jobs. Those people stayed. "Front-liners" and the engineers who supported them are the cut cohort.
2. The million-dollar salary bands go to survivors who ship with agents, not to the mid-layer generalists who built the product surface area between 2020 and 2024.
3. Gartner found ~80% of companies using autonomous tech have cut jobs but that those reductions are not translating into meaningful financial returns. Translation: Evans is betting on a thesis that has not paid out yet, and the engineers he let go know it.

That last point is a pitch, not just a data point. If you are hiring against ClickUp right now, "we ship faster with humans plus tools, not agents replacing humans" is the differentiated message, and Gartner is your credibility anchor.

The denial is the map. Every exec who insists it wasn't cost tells you exactly which sub-pool they just released.

The ex-ClickUp engineer pool, sized

In Refolk's index of professional profiles, there are 221 current and former ClickUp software engineers globally at the IC-through-Staff level, and only 6 of a 25-person sample are still at the company. The diaspora is already large, already scattered, and already warm to a second move.

Two things to notice in that number. First, at a 22% proportional cut across engineering, roughly 49 of those 221 are the freshest departures from this round. Second, the title mix skews senior: Senior Software Engineer, Software Engineer, and Staff Software Engineer show up 13, 6, and 5 times respectively in the sampled subset. Anyone still there in May 2026 had survived two prior culls, so the cohort's average tenure is 3 to 5 years and their product context is deep.

Here is how that stacks against the nearest comparable pool in project-management SaaS:

The Asana comparison matters because it tells you where the leverage is. Asana's pool is 1.56x larger but 3x more concentrated inside the mothership. Ex-ClickUp is the opposite: smaller headline pool, but most of it is already out, already interviewing, and already one warm intro away from moving. That is the pool you want a shortlist of by Friday, which is the exact gap Refolk closes. Describe the person in plain English ("senior full-stack engineer, 3+ years at ClickUp, shipped real-time collab or permissions, based in EMEA or US non-Bay") and get a ranked shortlist across GitHub, LinkedIn, and the open web without stitching boolean strings together.

Why the talent isn't in San Francisco

ClickUp's engineering footprint is a Kraków-and-Bengaluru story with a scattered US long tail, not a Bay Area cluster, and US-only outreach will miss roughly 40% of the pool. This is the single most under-priced fact about the cohort.

Of the top engineering regions in Refolk's index for ClickUp:

• Kraków, Poland is the largest single cluster.
• Bengaluru, India is the second.
• The US metros that appear are Portland, Pittsburgh, Richmond, and Greater Houston. Not San Francisco. Not New York.

Contrast Asana, where 9 of 25 sampled engineers are in San Francisco proper and another handful across the broader Bay Area, roughly 56% Bay-concentrated. If your sourcer's default filter is "United States" or "SF Bay Area," you will see maybe 60% of the ClickUp pool and none of the largest sub-cluster. For most PM-SaaS competitors that hire remote-first (Linear, Height, Coda, Airtable partially), the Kraków hub is the highest-signal, lowest-competition segment on the board this week.

What ex-ClickUp engineers actually shipped

Ex-ClickUp engineers ship the stack every PM-SaaS competitor needs: real-time collaborative documents, project views, spreadsheets, chat, permissions, and multi-tenant workspace infra, which maps directly onto Notion, Linear, Height, Coda, Airtable, and Monday. The transferability is unusually clean because ClickUp's product surface is one of the widest in the category.

Concretely, an ex-ClickUp staff engineer has likely touched:

• OT or CRDT-based real-time collaboration in documents and whiteboards.
• Rich-text editor internals (mentions, embeds, comments, suggestions).
• Multi-tenant workspace and permissions models at scale.
• Async workflow engines (automations, triggers, custom fields).
• Integrations layer against Slack, GitHub, Google Workspace, and the usual PM ecosystem.
• Mobile plus web plus desktop parity, which is rarer than it sounds.

That is a very specific technical fingerprint, and it is what makes boolean sourcing so bad at finding this pool. "ClickUp" as a keyword pulls users, consultants, and integration partners as much as it pulls engineers. Describing what someone built in plain English ("engineers who shipped CRDT-based document collab in a multi-tenant SaaS") is the only way to filter down to the 221 that actually matter, and it is another place Refolk does the work that boolean cannot.

The outreach angle that actually lands

The message that lands with this cohort is not "we're hiring," it is "we ship with humans, and here is what your ClickUp context is worth in dollars." Two independent signals say reply rates should be elevated.

First, the Glassdoor pattern. Public reviews from the layoff period include lines like "immature and out of touch leadership, multiple rounds of layoffs in the past couple of years and offshoring more and more roles, the platform itself is buggy and unreliable." That is the trailhead. Engineers who type that into Glassdoor also open cold email.

Second, the credibility wedge. Evans went public with an unusually explicit AI-substitution frame; most of the industry (Meta cutting 8,000 the same week on record revenue, Oracle eliminating up to 30,000 to fund AI infrastructure, GitLab restructuring for the "agentic era") is doing the same in softer language. Ex-ClickUp engineers have watched the pattern from the inside. A pitch that says the quiet part out loud, that human engineers with good tools still ship faster than agent swarms in production SaaS, differentiates immediately.

Three concrete outreach moves for this week:

1. Split the pool into "cut in May 2026" and "left in prior rounds." The former needs urgency and specificity. The latter needs a reason to move again.
2. Lead with the product surface, not the company. "You shipped the ClickUp permissions rewrite" beats "we saw you were at ClickUp" every time.
3. Anchor comp against the $1M-band framing. Evans made money the story. You can either match the frame ("here is our senior band and equity math") or reject it ("we pay in cash and shipped software, not agent-supervision premiums"), but you cannot ignore it.

221

FAQ

How many engineers did ClickUp actually cut in the 2026 round?

ClickUp cut about 285 to 290 people total, or 22% of a ~1,300-person workforce, announced by CEO Zeb Evans on May 21, 2026. Public reporting does not break out engineering versus CX versus GTM cleanly, but if engineering was cut proportionally, roughly 49 of the 221 ClickUp software engineers in Refolk's index are part of that cohort. The remainder of the ex-ClickUp pool left in the 2022 (7%) and 2023 (10%) rounds or on their own between rounds.

Why does the "not cost cutting" framing matter for sourcing?

Because it tells you which sub-pool was cut. Evans explicitly separated the future workforce into "builders," "system managers," and "front-liners," and said savings would flow into $1M salary bands for the first two groups. The people who left were disproportionately the non-AI-native mid-layer engineers and CX-adjacent tooling teams. Sourcing "ex-ClickUp AI engineers" is chasing the wrong end of the pool. The right end is real-time collab, workflow, permissions, and integrations engineers.

Where should we source ex-ClickUp engineers geographically?

Kraków, Poland first, Bengaluru second, then a scattered US long tail across Portland, Pittsburgh, Richmond, and Greater Houston. Notably not San Francisco or New York. Competitors running US-only or Bay-only searches will miss the largest single cluster of the pool. If you are remote-friendly and can hire in EMEA, Kraków is the highest-signal, lowest-competition segment on the board right now.

Which PM-SaaS competitors have the best chance of intercepting this pool?

Any competitor whose product surface overlaps with ClickUp's stack (docs, tasks, spreadsheets, chat, permissions) can absorb this talent cleanly: Linear, Notion, Height, Asana, Monday, Airtable, Coda, Smartsheet, Wrike, and Basecamp. The differentiator is not brand, it is the outreach message. Teams pitching "humans plus tools, not agents replacing humans" and citing Gartner's finding that 80% of agent-driven cuts have not produced financial returns will out-convert teams leading with generic comp pitches.