Yann LeCun

Why Yann is the canonical Working Researcher

Yann LeCun spent eight years at AT&T Bell Labs developing the foundational architecture of modern AI. Working in the Adaptive Systems Research Department from 1988, he built convolutional neural networks (CNNs) — a biologically inspired approach to image recognition that borrowed from the neuroscience of the visual cortex. The practical result was LeNet, a system that could read handwritten digits with sufficient accuracy to be deployed at scale: by the late 1990s it was processing roughly ten percent of all checks deposited in the United States. This was not a demo. It was production AI, running in banks, years before the term 'deep learning' existed.

In 2003 he joined New York University as a full professor at the Courant Institute of Mathematical Sciences, where he has held the Jacob T. Schwartz Chair in Computer Science since 2023. A decade later, in December 2013, he became the founding director of Meta's Fundamental AI Research lab (FAIR), a position he occupied for over a decade while retaining his NYU professorship — the dual appointment itself a statement about what the role is. He did not leave academia to run a startup. He ran a corporate research lab the way a university scientist runs a department: with a publication mandate, peer-reviewed output, and an identity rooted in the field rather than in the company. In late 2025 he left Meta to co-found AMI Labs, focused on world-model architectures.

The Turing Award LeCun shared in 2018 with Yoshua Bengio and Geoffrey Hinton — computing's highest honor, often called the 'Nobel Prize of computer science' — was a collective recognition of thirty years of institutional research that the industry initially dismissed. LeCun, Bengio, and Hinton continued publishing through two AI winters, sustained by university positions and industrial lab salaries rather than by investor conviction. That is the Working Researcher's defining move: advancing the field through the peer-reviewed record and the institutional apparatus, not through a founding equity stake. LeCun did not build the companies that monetized deep learning. He built and published the foundations that made those companies possible — and collected his Turing Award when the field finally caught up.

The Working Researcher salaried researcher inside an organized research function — publishes, wins grants, builds citation impact. See the Working Researcher role →

Same role, different wave