Block's 40% Cut: The Real Ex-Block Engineer Pool Is 1,800, Not 2,800

Block announced on Feb 26, 2026 that it is cutting more than 4,000 employees, roughly 40% of a 10,205-person workforce, effective immediately per Jack Dorsey's staff memo. A Blind thread claiming "70% of engineering was laid off" started circulating within the week and is already driving founder Slack channels into a frenzy. Before you queue up 300 InMails, understand that the number is almost certainly inflated, half the resumes you'll see are performance-managed, and the window to reach the engineers you actually want is shorter than the window to reach everyone else.

Why the "70% of engineering" claim doesn't reconcile

The 70% figure is an anonymous Teamblind post, not a Block disclosure, and the math falls apart against the total headcount cut. Block is going from 10,205 employees to "just under 6,000," per Engadget's Feb 26 coverage. That is roughly 4,200 net departures.

Run the arithmetic:

• If engineering is 40% of Block's headcount (a generous fintech-standard assumption), that is about 4,080 engineers pre-cut.
• 70% of 4,080 is about 2,860 engineering cuts alone.
• That leaves only about 1,340 cuts distributed across Square's field sales org, compliance, support, ops, and G&A at a company famous for its physical Square merchant footprint.
• Block is not gutting Square sales while keeping payments engineers. The distribution is implausible.

A more defensible read: engineering cuts are proportional or slightly heavier than the 40% company average. That puts the real hireable engineer pool at 1,400 to 1,800, not 2,800+. That matters because your sourcing funnel, your comp bands, and your "how urgent is this" gut check should all be sized against the real number.

1,400 - 1,800

The 14-day window is real, but only for the top 15%

The urgency window applies to AI-adjacent pivoters, not pure payments-domain engineers, and confusing the two will waste your first two weeks. Tech-Insider's post-cut labor read is that workers finding new roles fastest are pivoting into MLOps, AI safety, prompt engineering, and data infrastructure. Those profiles move in days.

Payments-rails, PCI-scope, and ledger engineers move slower for a structural reason: there are only so many jobs that require the exact skill.

Refolk's index currently shows 427 US-based engineers (IC and EM level) tagged with Payments skills. That is the entire adjacent competitive pool you are fighting Block alumni for. Fintechs run slow interview loops on this cohort, so a specialist coming out of Cash App's ledger team can sit for 60 to 90 days without a signed offer.

Practical sequencing:

1. Days 0 to 14: hunt Goose contributors, Cash App AI/ML, and TBD/Spiral engineers pivoting toward agent tooling. These get poached first.
2. Days 14 to 60: work the payments-rails specialists deliberately. They are not going anywhere.
3. Days 60+: pick up the Afterpay BNPL engineers who took severance-fueled sabbaticals and are returning to the market.

What the ex-Block engineer pool actually looks like

Block's cuts hit the same three orgs as the January 2024 round: Cash App, foundational, and Square, per Fintech Futures' coverage of Dorsey's memo. Layer in the Bitcoin-adjacent subsidiaries (TBD, Spiral, Bitkey), which are disproportionately exposed because Block's Bitcoin revenue has fallen with the coin price, and you get a fairly clean org-by-org picture.

Here is what the surrounding competitive pool looks like when you strip it down:

The 1.8:1 senior-IC-to-EM ratio in the adjacent pool is the number to internalize before you open your inbox on Monday. Block's March 2025 cut eliminated 80 manager positions specifically to "flatten" the hierarchy, per Payments Dive. Dorsey has done it again. You will get a wall of Engineering Manager and Senior EM resumes in the first week. Most of them are not what you need.

Half the resumes you'll see are performance-managed

Roughly half of the ex-Block resumes hitting your inbox are performance-driven cuts, not strategy cuts, based on Block's own disclosure pattern. In the March 2025 round, of 931 jobs eliminated, 460 were cut for performance reasons and 391 for strategy, per Payments Dive. That is 49.4% performance-driven.

Dorsey has bundled performance management into strategic layoff waves in every prior round. There is no reason to think this Feb 26 cut is different. The mechanism is straightforward: it is cheaper, faster, and legally cleaner to fold a stack-rank quartile into a headcount reduction than to run individual PIPs to completion.

What this means for sourcing:

• LinkedIn "Open to Work" scraping oversamples the bottom half of Block's distribution. These are the profiles that were going to be cut anyway.
• Referral chains from current or recent ex-Block engineers filter better than any keyword search. Ask the good ones who they'd hire.
• A domain screen matters more than a resume screen. Anyone who ran a real ledger migration, shipped PCI-scope changes, or on-called for Cash App's payment authorization service is signal. Title alone is noise.

This is the exact gap Refolk closes for a founder trying to move in a week. You describe the person in plain English ("staff engineer who worked on Cash App's ledger or authorization systems, not a manager, not on PIP watch"), and Refolk returns a ranked shortlist pulled across GitHub, LinkedIn, and the open web with the domain signals attached.

The highest-signal subsets: Goose, TBD, and Afterpay

Three specific product surfaces at Block produced engineers worth breaking your interview calendar for: codename Goose, the Bitcoin subsidiaries, and Afterpay. Each maps to a specific type of buyer.

Codename Goose

Block built an internal LLM agent called "codename goose" for interacting with LLMs, confirmed by Engadget. Ex-Block engineers who touched Goose are the single highest-signal subset for anyone building agentic developer tooling, IDE copilots, or agent runtimes. They have done the unglamorous part: tool-use plumbing, sandboxing, eval harnesses, and prompt/response logging inside a compliance-heavy enterprise.

If you are building anything in the agentic dev-tool adjacency, Goose alumni are the poach.

TBD, Spiral, and Bitkey

Block's Bitcoin-adjacent orgs are disproportionately exposed to every cut because Block's Bitcoin revenue moves with the coin price. These engineers ship self-custody, Lightning, and on-chain settlement code. The natural buyers are Coinbase, Kraken, Anchorage, and stablecoin-rails startups. Sourcing angle: crypto-native profiles rarely surface cleanly on LinkedIn because their public work lives on GitHub and in protocol repos. A LinkedIn-only search will miss most of them.

Afterpay engineering

Block acquired Afterpay in 2021 for $29 billion. The BNPL engineering org is prime for Klarna and Affirm poaching. Merchant-integration engineers from Afterpay are also strong candidates for any commerce startup building checkout, orchestration, or fraud.

The 14-day window applies to Goose and MLOps profiles. Payments-rails specialists will still be available in September.

NYC is the arbitrage, not SF

NYC leads SF Bay Area 1.67 to 1 for payments engineers in the adjacent pool, which contradicts the default founder instinct to post everything in SF. In Refolk's index, NYC shows 5 top-region matches vs SF's 3 for the "Payments" skill tag. Most ex-Block engineers are Bay-based (Block is San Francisco-headquartered), which means the negotiating leverage sits with NYC-headquartered fintechs.

If you are at Ramp, Mercury, Rain, Alloy, or any NYC-based payments company:

• Pitch "no relocation, denser payments engineering community, better founder access" and mean it.
• Block's Bay Area engineers with kids in school will not move. Block's Bay Area engineers under 32 will consider it. Segment your outreach accordingly.
• If you are SF-based and hunting Block alumni, your competition is Meta (about 28% of the adjacent pool), which will simply outbid you on cash. Compete on scope and product surface, not comp.

One more reason not to hedge: Block's stock spiked 20% in after-hours trading on the layoff news, per Forbes' Feb 27 coverage. The market cheered, which means Block has zero incentive to reverse or slow-roll cuts. Unlike the 2023 tech-layoff cycle where boomerangs were common, you can make an offer without worrying the candidate will bail for a re-hire.

The 14-day sourcing playbook

Move on Goose and AI-adjacent profiles first, use referral chains not scraping to avoid the performance-cut oversample, and do not burn the two weeks on generalists who will still be available in October. The compressed version:

1. Day 1 to 3: Pull a targeted list of Block engineers with GitHub activity on Goose, MLOps, or agent tooling in the last 12 months. Skip anyone whose most recent public commit is older than six months.
2. Day 3 to 7: Warm-intro through your best current engineer who has an ex-Block colleague. One warm intro beats fifty cold InMails on this cohort.
3. Day 7 to 14: Send targeted messages to TBD/Spiral engineers if you are crypto-adjacent, and to Afterpay engineers if you are commerce-adjacent. Named product surface in the first line, always.
4. Day 14 to 30: Payments-rails specialists get slower, more thoughtful outreach. These interviews should be scoped conversations, not pipeline pushes.
5. Day 30+: Circle back to the sabbatical returners. Severance from a 40% cut buys 3 to 6 months of runway, so a lot of good engineers will not respond to anything in the first month by choice.

Refolk speeds up steps 1 and 3 specifically. Instead of building Boolean strings or paying an agency retainer to filter, you ask in plain English ("ex-Block IC engineers with Bitcoin protocol commits in the last year, US only") and get back a ranked shortlist with the underlying evidence. That is the difference between reaching 20 of the right engineers in week one and reaching 400 of the wrong ones.

FAQ

Is the "70% of engineering laid off" Blind claim accurate?

Almost certainly not. The number originates from an anonymous internal engineer on Teamblind and is not corroborated by Block. The arithmetic fails: 70% of engineering would represent about 2,860 of Block's roughly 4,200 total cuts if engineering is even 40% of headcount, leaving implausibly few cuts for Square's large field sales org, compliance, and support. A more defensible estimate is 1,400 to 1,800 engineers cut, roughly proportional to the 40% company-wide reduction.

How much of the ex-Block resume flood is performance-managed?

Around half, based on Block's own precedent. In the March 2025 round, 460 of 931 cuts (49.4%) were performance-based, per Payments Dive. Dorsey bundles performance cuts into strategic layoff waves in every round, so any list scraped from LinkedIn "Open to Work" filters will oversample the bottom half of Block's distribution. Referral chains and domain-specific screens filter far better than resume keywords.

Which ex-Block engineers should I prioritize in the first two weeks?

The engineers who touched Block's Goose agent, MLOps and data infrastructure roles, and TBD/Spiral/Bitkey Bitcoin engineers. These profiles move fastest because AI-adjacent pivoters are getting hired within days and crypto-native engineers have a dozen well-funded landing spots. Pure payments-rails and PCI-scope specialists move slower (60 to 90 days) because there are only about 427 US roles in the adjacent pool, so you can afford to work those candidates thoughtfully.

Should NYC fintechs be more aggressive than SF fintechs on this pool?

Yes. Refolk's index shows NYC leading SF Bay Area 1.67 to 1 for payments engineers, and NYC-headquartered fintechs like Ramp, Mercury, Rain, and Alloy can credibly pitch "no relocation, denser payments community" to the ex-Block engineers open to leaving the Bay Area. SF-based buyers compete directly with Meta (about 28% of the adjacent employer share), which will win on cash. NYC buyers compete on scope and product surface, which is a fight they can actually win.