Create the Conditions for Life's Work

The Jensen frame

The principle originates with Jensen Huang and is relayed and adopted by Eric Glyman, who calls it "a bit inverted" but "really profound." Huang, in Glyman's account, "believes his role as CEO is to create the conditions by which people can do their life's work."¹ Glyman breaks it into two parts.

The first is to create the conditions: find extraordinary people and build an environment where they want to work and feel motivated, "the right talent in seat, the right culture." The second is the phrase "do their life's work," which shifts the CEO's leverage to work selection once people of that caliber are hired: "which work do I select and which do I deselect from the company?" Huang's own example is that Nvidia could have safely grown from "a $5B to a $10B or $20B company" by going deeper in video-game chips, but he repeatedly redirected the company toward GPUs for artificial intelligence and genomic research because that was the more important work. Huang assesses his own effectiveness partly by asking, "do you select the right work?"

Glyman's pairing: everyone is the hero of their own story

Glyman fuses the frame with a lesson he traces to Capital One: interviews over-index on skills and leave out motivation. "Everyone is the hero of their own story, understanding people's motivations, what they want to do, where they want to be in years, and periodically referencing back." Concretely, someone who wants to start a company one day gets placed on zero-to-one problems, and the check-in comes when they reach "one" about whether they want to go one-to-ten. Because "companies are a set of people," people progressing against their own goals is what he argues gives an organization longevity. He names Karim Atiyeh, Jeff, and Diego as leaders particularly strong at this.

The operational consequence is where Glyman spends his time. He estimates roughly a third on hiring, plus more time inspecting whether smart people are producing, whether people are blocking each other, and whether the organization is efficient. He reserves his unique levers, "the power to convene meetings, change structures in organizations," for what only he can do.

Trust, tenure, and free-riding

Retelling the frame with David Senra, Glyman restates it nearly verbatim, "my job is to create the conditions for people to do their life's work, to unblock these other people," and adds that he does not believe he is the smartest person at Ramp.² He extends it in two directions. Trust compounds into throughput: he wants to work with people early and for a long time, citing the shorthand that Buffett and Munger reached where they no longer had to call each other, and David Senra's Mr Beast example, where operations lead Tyler already knows the shot before Mr Beast speaks, so that "in every single hour the throughput that organization gets is radically higher." And free-riding is the corrosive failure: people with similar pay and ownership who do not pull their weight break the belief that the organization respects high standards, the belief the whole conditions frame depends on. The hero-of-their-own-story lesson recurs here as a hiring filter as much as a management move: find a candidate's actual motivation independent of Ramp, and if no evidence suggests it coincides, "don't waste your time."

Where it fits

The principle reframes the CEO from operator and decider to environment-and-portfolio designer, closer to designing the environment than to command-and-control. The select-and-deselect half is the companion to team-level autonomy: small teams own the how while the CEO chooses which problems exist. The who the conditions are built for is the extraordinary, spiky person described in hire for spikes and the drive-first profile of the determined generalist, and the quality of those people is the recurring premise of the A-player framework. The open tension Glyman leaves is that selecting the right work is high-variance, legible mainly because Huang's company-gambles won, and that a person's preferred work may not be the work that needs doing, which he frames as a longevity input rather than an override.