Debunking the 30 Total Engineers Hype at X

The recent viral claims about X running on just 30 engineers have sparked a lot of celebration in certain circles, posts framing it as proof that massive layoffs + extreme leanness = superior efficiency.

But let’s challenge the terminology and add some missing context before everyone rushes to declare small-team supremacy the new universal truth on X.

What Nikita Bier actually said (Jan 10, 2026):

“We only have ~30 engineers on the product but they’re the best in the business” — @nikitabier, replying to @BillyM2k and @BoredElonMusk

Note the qualifier: on the product. This refers to the core product engineering group, the slice directly building/iterating on user-facing surfaces like the feed, onboarding, interactions, Grok integrations, etc. It’s not a claim of 30 engineers total across the entire company.

Raman Khatri echoed this vibe (Feb 12, 2026):

“I’ve personally called nikita on a Tuesday night to get his green light on massive projects that were shipped by Friday afternoon. That same project would’ve taken a year at most other companies.” — @ramankhatri, quoting Nikita’s team-size meme

“Team Size — Meta: 87,000, Microsoft: 221,000, Google: 190,000, … X: 30” — @nikitabier

Again, that’s describing the velocity of a small, trusted, decision-empowered squad, not the full org required to operate a global platform.

The broader reality (based on public statements and patterns):

Pre-2022 Twitter had ~1,500 engineers. Post-acquisition cuts were deep (down to low hundreds total staff by 2024–2025 reports).

Today, the product-facing engineering core is indeed tiny (~30 as stated by Nikita), which explains the fast iteration on visible changes.

But keeping X running involves more: infrastructure/scaling, ads backend/monetization rails, trust & safety moderation pipelines, payments/fraud (especially with X Money prep), compliance/legal for global markets, security, data infra, AI rec systems (Elon has mentioned separate ~30 xAI folks on that in various updates).

“There are another 30 or so xAI engineers working on the pure AI-based recommendation system” — @elonmusk, replying to @DallasAptGP

The total engineering headcount is higher than 30, still dramatically smaller than legacy Twitter or any FAANG, but not literally 30 people holding everything together.

On “massive projects” vs. features:

The shipped examples often celebrated as evidence (video reactions, Grok widgets/voice, smart cashtags, profile transparency tweaks, algo boosts for friends content, draft sync, starterpacks, AI spam crackdowns, manipulated media labels) are mostly meaningful product features, UI/policy iterations, or quick policy/enforcement changes. They move fast because of minimal process overhead, fair flex for velocity.

But “massive projects” in Raman’s post implies bigger scope: full architectural rewrites (e.g., search/ads engine from ground up), new standalone products (X Money payments infra, X Chat spin-out), major monetization overhauls, or compliance-heavy systems. Those teased items (X Money testing, niche communities refresh, or creator tools hub) are in progress but not “shipped by Friday” territory. The hype often blurs iterative speed with full-scale project delivery.

For people cheering “30 total engineers”:

It’s understandable to celebrate leanness after years of bloated tech orgs (some posts even list “30 engineers, 2 designers, 1.5 product managers + @nikitabier” as the full team).

But the narrative jump from “core product team of ~30 ships fast” → “entire company runs on 30 people total” ignores the supporting layers still needed.

X isn’t a pure 30-person startup. It’s a high-traffic, regulated, global platform with ongoing infra/security/compliance demands. The real story is how a dramatically reduced org (with a hyper-empowered core squad) sustains operations and iterates quickly — not that 30 humans alone maintain uptime, fight spam at scale, handle payments fraud, etc.

If the goal is critiquing FAANG bloat, fine. But let’s be precise with the claims: it’s a lean product engineering core enabling startup-like speed on visible changes, not a literal 30-person skeleton crew running the whole show.

What do you actually count as a “massive project” that proves the model? Algo/arch rewrite? Full payments launch? Curious where the skepticism lands.