How Claude Code Became Anthropic’s $2.5 Billion Bet on AI at Work
November 15, 2026
After nearly seven years at Meta, where he worked across Instagram, Facebook Groups and the code quality team, Boris Cherny joined Anthropic in 2024 and began experimenting with ways to let Claude interact with a computer.
A month of testing Anthropic’s models led Cherny to build a command-line prototype that gave Claude access to a computer’s operating system, laying the groundwork for Claude Code.
Cherny, Anthropic’s head of applications, developed Claude Code into an AI agent that can write code, run commands and access files; its success became a major driver of workplace adoption and company growth.
Claude Code surpassed $2.5 billion in annualized run-rate revenue in February, making it important to Anthropic’s commercial ambitions and anticipated IPO, though the company does not publish usage figures.
After Claude Code became generally available in May 2025, enterprise AI spending accelerated, according to Ramp economist Ara Kharazian, who described coding agents as an early commercially advanced AI use that directly boosts productivity in specific jobs.
The tool’s reach has extended beyond software: staff at the Mid-Ohio Food Collective used it to build a game that helps families choose healthier foods, advancing an idea first discussed in 2017 that the organization had not had time to pursue.
Seeing businesses use Claude Code for nontechnical work prompted Anthropic to create Claude Cowork, an AI agent for office tasks; the company built it in 10 days using Claude Code.
Cherny says close contact with users is central to improving the products, and he regularly gathers feedback directly, including by responding to Claude users on X.
Cherny expects AI to take over many coding tasks and predicted software engineering roles could begin disappearing by the end of 2026, as workers shift toward broader “builder” responsibilities such as collaborating across teams and deciding what to make.
But many programmers remain wary: a Stack Overflow survey of 30,000 programmers found that six in 10 technology workers actively try to avoid AI, citing doubts about its capabilities and concerns about added pressure and burnout.
Anthropic is developing tools that automate work while warning about AI risks, and it faces competition from OpenAI, reliance on a limited number of customers and pressure to justify its high valuation.
Cherny says he cannot predict how quickly engineering work will change and values patience and long-term thinking, lessons reflected in his hobby of making miso, which he learned while living in Japan.
Summary based on 4 sources
Get a daily email with more Tech stories
Sources

CNN • Nov 15, 2026
Meet the miso-making engineer behind Anthropic’s biggest hit
NewsBytes • Nov 16, 2026
Meet Boris Cherny, former Meta engineer behind Anthropic's Claude Code
KVIA • Nov 15, 2026
Meet the miso-making engineer behind Anthropic’s biggest hit
SSBCrack News • Nov 15, 2026
How Boris Cherny’s Claude Code Became Anthropic’s Breakout AI Product—and Could Reshape Engineering Jobs -