Arman Hezarkhani, co-founder of 10x, presents a radical thesis: in the AI era, the traditional engineer compensation model (fixed salary or hourly rate) is obsolete because it fails to reward the exponential productivity enabled by modern tools. He observes a fracture between engineers who use AI extensively (orchestrating dozens of agents) and those who still code "like cavemen," character by character. For him, this is an incentive problem: why go 10 times faster if you're paid the same?
10x's solution is to pay engineers like salespeople, meaning based on performance, tied to actual output. Concretely, they use a system based on Story Points. The workflow is as follows: 1. A Strategist (a distinct role, compensated on customer satisfaction/NRR) defines the architecture and breaks the project down into tickets, assigning a point value to each. 2. The Engineer executes the tickets. 3. Once the ticket is validated by the Strategist and the Client (double quality control), the points are earned. 4. Compensation includes a fixed base and a significant variable component indexed on the volume of points delivered.
Hezarkhani addresses the obvious criticisms (risk of declining quality, point inflation) through the role structure: the Strategist acts as a counterweight to the Engineer for scoping and validation. According to him, this model attracts "10x engineers" (former founders, elite researchers) who know they can produce enormously thanks to AI and want to capture a share of the value created, rather than waiting for a hypothetical exit through stock options. He concludes that AI grants superpowers, but that without an aligned compensation system, employees have no reason to use them to their full potential.