Uber CTO says it's going big on AI coding: 'This is a real reset moment for engineering'

Uber, renowned for revolutionizing ride-hailing, is now poised to transform its software development process through artificial intelligence. Praveen Neppalli Naga, Uber's Chief Technology Officer, revealed in a recent LinkedIn post that an impressive 95% of the company's engineers are utilizing AI on a monthly basis, marking what he calls a significant 'reset moment for engineering.' Over recent months, Uber has aggressively embraced AI coding, and Naga noted that the outcomes have been remarkable. The most significant development, he explained, comes from agentic AI—technology capable of performing tasks independently with minimal oversight. This paradigm shift has allowed Uber's engineers to increasingly hand over responsibilities to AI, rather than merely following its suggestions. As a result, the company is witnessing an astonishing 1,800 code changes weekly being generated entirely by its internal AI coding agent. Naga highlighted that the contribution of this agent has surged from under 1% to 8% of all coding modifications. 'There is zero human authoring,' he stated, explaining that while engineers review and approve the code, the actual writing is done solely by AI. Uber's initiative aligns with broader trends in the tech industry. A study by engineering intelligence platform Jellyfish revealed that 63% of 700 surveyed companies are now employing AI tools for a majority of their coding tasks. Some businesses are even incorporating AI tool usage into employee performance evaluations to encourage adaptation. However, Naga emphasized that at Uber, the most significant adoption comes from engineers who are 'quietly experimenting' with these technologies, rather than it being mandated from the top levels of management. Despite the advantages, the integration of AI into coding processes has not been without its challenges. For instance, Amazon recently implemented new safeguards after a series of outages, one of which was largely attributed to its AI coding tool and resulted in the loss of nearly 120,000 orders. This rapid evolution in AI coding is causing considerable upheaval in the field of software engineering, traditionally viewed as a secure and lucrative career. Some engineers have reported that AI tools are actually increasing their workload, leading to feelings of burnout. 'The role of the engineer is shifting—from writing every line to architecting systems and reviewing AI-generated code,' Naga concluded, highlighting the changing landscape of the profession.