Ex-Meta Manager: 2% of Engineers Drive Major AI Productivity Gains

Kun Chen, a former engineering manager at Meta, said that only 2% of engineers are using AI very effectively, resulting in outsized productivity gains for that small group. Chen, who previously worked as an engineer at Microsoft and Atlassian, shared these insights during an episode of Steve Huynh's 'A Life Engineered' podcast, released on Monday.

He emphasized that companies will allocate the most impactful projects to this 2% of engineers who leverage AI effectively.

Business Insider reported that Chen, based on conversations with chief technology officers in his most recent role, observed most companies experiencing a 10% to 15% productivity boost from AI overall. He attributed this modest gain to the majority of employees using AI in a shallow way, contrasting with the 2% who achieve massive shifts in their work.

'When these CTOs zoom in, what they see is that in their company there is maybe 2% of people who actually figured out how to use AI very effectively,' Chen said.

He added that mastering agentic engineering leads to a massive boost in productivity, but this is only achieved by a small number of developers. Chen stated that only about 2% of engineers are getting outsized results from AI. He described how companies are reallocating the most impactful projects to this 2%, allowing them to execute faster while larger teams lag on minor tasks.

This group experiences a massive shift in how they work, prompting firms to double down on them amid industry changes. Chen said the 2% of engineers are experiencing a massive shift in how they work.

He urged developers to adopt a mindset of continuous learning to adapt to AI's rapid evolution, rather than focusing on specific tools that may become outdated. 'We should invest time in a different mindset of continuous learning,' Chen said. He compared AI's trajectory to past revolutions like the industrial era or the internet, starting small before widespread adoption.

Business Insider reported Chen's comments come as tech companies like Meta and Amazon conduct layoffs and reorganize into smaller teams. He noted an urgency for engineers to figure out effective AI use to avoid falling behind. 'Otherwise they may indeed fall behind,' Chen told Business Insider.

He highlighted that growing the 2% requires collective efforts in education and awareness within the engineering community. Chen said companies are reallocating the most impactful projects to the 2% of engineers using AI effectively. At some point, CTOs will question the need for slower teams, he added, potentially leading to further reallocations.