Kilby Says: Jensen Huang Knows What He's Doing
My exchange with Kilby about NVIDIA founder and CEO Jensen Huang’s leadership style. We cover Kilby’s leadership library.
Me: My understanding is that Jensen Huang - NVIDIA’s CEO - has 60 direct reports. What kind of CEO spreads himself so thin?
KILBY: Jensen’s 55–60 direct reports isn’t thin — it’s the point. He’s running a fundamentally different organizational architecture, and it’s one of the most deliberate leadership designs in tech.
The model has two interlocking pieces:
1. The Information Broadcast Model. Huang holds zero regularly scheduled one-on-ones. All strategic feedback is delivered in group settings so every executive hears the same reasoning at the same time. His logic: when information cascades through management layers, each layer filters and distorts it. By broadcasting to everyone simultaneously, he eliminates the telephone game, kills information hoarding as a power lever, and compresses the time it takes for a decision to reach the people who need to act on it. The trade-off is explicit — no individualized CEO mentorship. That’s acceptable because NVIDIA’s leaders are expected to execute autonomously.
2. Mission Is the Boss — Pilot in Command. For each strategic mission, one person is designated the “Pilot in Command” (PIC) with full accountability and a direct line to Huang. The PIC pulls cross-functional resources from anywhere in the organization. Job titles reflect functional expertise, not rank. The org reconfigures around missions, not around reporting lines. Huang describes it as “wired up like a neural network” — connections form and dissolve based on what the mission requires.
Why it works at NVIDIA specifically:
Accountability is unambiguous. One person owns each mission. No shared accountability, no finger-pointing at organizational seams.
Resources flow to priority, not politics. The highest-priority mission gets the best people from anywhere, because the mission outranks the org chart.
Speed of reconfiguration. Traditional reorgs take months. Mission reassignment takes days — the underlying network of people stays stable while the mission overlay shifts.
The prerequisite most companies can’t meet: People must derive status from missions they lead and outcomes they deliver — not from headcount or title. In organizations where the currency of prestige is team size and title inflation, this model collapses because leaders resist releasing resources to other missions.
The honest assessment: This is not a CEO spreading himself thin. It’s a CEO who decided that the bottleneck in most organizations is information fidelity and resource allocation speed — and designed a structure that optimizes for both at the cost of conventional management hierarchy. It requires a CEO with extraordinary bandwidth and a culture that rewards output over org-chart position. Huang has both. Most CEOs don’t, which is why almost nobody runs this way.
The results speak for themselves — $81.6B in quarterly revenue with 75% gross margins, executing across hyperscale, enterprise, sovereign, and robotics simultaneously. The organizational design isn’t incidental to that execution. It’s enabling it.
Me: Does Jensen’s leadership style intersect with any of our Leadership frameworks?
KILBY: Jensen’s leadership style intersects with at least five distinct frameworks in the library. Here’s how they map:
1. The Information Broadcast Model — This is Jensen’s own, and it’s the most structurally radical. 55–60 direct reports, zero scheduled one-on-ones, all strategic feedback delivered in group settings. The design eliminates information distortion across management layers, kills information hoarding as a political lever, and compresses decision propagation from weeks to minutes. The trade-off — no individualized CEO mentorship — is accepted because leaders are expected to execute autonomously.
2. Mission Is the Boss — Pilot in Command — Also Jensen’s. The org reconfigures dynamically around missions, not reporting lines. One person owns each mission with full accountability and a direct line to Huang. Resources flow to priority, not to whoever has the biggest team. This is what makes the flat structure functional — without it, 60 direct reports would be chaos. The PIC model gives it shape.
3. Top-5-Things (T5T) Email System — The third piece of Jensen’s information architecture. Any employee can send Huang a short email listing their five most important priorities or challenges. He reads ~100 of these daily. This gives him unfiltered ground-truth signal that bypasses the sanitization of normal reporting chains. Critically, he does not send his own T5T — if he did, people would start mimicking his priorities and destroy the signal value.
4. Suffering as Character Formation — Jensen’s talent philosophy. He prefers to push people past their comfort zone rather than fire them, views near-death experiences (like NVIDIA’s 1996 Sega crisis) as character-building for the institution, and hires for resilience over raw intelligence. This pairs with his Thirty Days from Bankruptcy operating posture — maintaining the emotional urgency of survival even at $4.8 trillion in market cap.
5. Horowitz’s Ones and Twos Framework — This is where an external lens helps. Jensen is the purest “One” in the framework — he sets direction, makes decisions on incomplete information at extraordinary speed, and derives energy from strategic complexity. Horowitz observes that most great One CEOs staff their teams with Twos and Functional Ones. Jensen’s flat structure takes this further: because every leader reports directly to him, there’s no ambiguity about who the One is. The Pilot in Command model gives Functional Ones room to operate autonomously in their missions without creating strategic contention at the top.
The meta-insight across these frameworks: Jensen’s system isn’t a collection of separate ideas — it’s an integrated architecture. The flat structure (Broadcast Model) eliminates information distortion. The PIC model provides accountability and resource allocation without hierarchy. T5T provides unfiltered ground-truth. The suffering philosophy produces people who can operate under the pressure the system creates. And the 30-days-from-bankruptcy posture ensures the whole thing stays urgent.
Horowitz’s confidence curve framework also explains why Jensen can run this system and most CEOs can’t. Horowitz points out that Jensen has been doing this for 30+ years — he sits at the far end of the CEO confidence curve where decisions are fast, first-principles, and uncontaminated by political calculation. That level of judgment is what makes a 60-person span of control work. Without it, the system collapses.