Uber’s CTO Praveen Neppalli Naga has declared that the company has ‘leaned in hard’ to AI coding, with remarkable results. According to a LinkedIn post, 95% of Uber’s engineers now use AI tools every month, making 1,800 code changes weekly through their AI agent. This widespread adoption represents what Naga calls ‘a real reset moment for engineering,’ highlighting the fundamental shift occurring in how software is created across the industry.
The most significant transformation comes from agentic AI—systems that complete tasks autonomously with minimal supervision. This paradigm shift means engineers are moving from writing every line of code to focusing on system architecture and reviewing AI-generated outputs. Naga emphasizes that at Uber, the strongest adoption is emerging from engineers ‘quietly experimenting’ rather than being mandated top-down. This bottom-up approach suggests a more organic integration of AI tools into existing workflows, potentially leading to more sustainable adoption patterns.
However, the rapid advancement of AI coding tools presents both opportunities and challenges for the software engineering profession. While some engineers report increased efficiency, others note they’re working harder, potentially leading to burnout. The case of Amazon is particularly instructive, as they had to implement new guardrails after outages linked to their AI coding tool resulted in nearly 120,000 lost orders. As the industry continues to evolve, the key question emerges: how can organizations harness AI’s power while maintaining code quality and preventing over-reliance on automated systems?
