TLDR: In a recent TEDxCSTU talk, veteran technologist Raymond Fu addressed the evolving landscape of software engineering in the age of Artificial Intelligence. He asserted that despite AI’s growing coding capabilities, human software engineers remain essential, with their roles shifting from mere coders to visionaries, system architects, and ethical leaders. Fu emphasized the importance of critical thinking, system design, ethical considerations, and collaborative work with AI for the next generation of engineers.
Veteran technologist and entrepreneur Raymond Fu recently delivered an insightful TEDxCSTU talk, originally held on May 18, 2025, at the California Science & Technology University (CSTU), addressing a pivotal question for the tech industry: ‘If AI can write code, is it still worth learning software engineering?’ Fu’s unequivocal answer is yes, asserting that the profession remains profoundly essential, albeit undergoing a significant transformation.
Fu, drawing from decades of experience in leading innovation and his roles as a professor and parent, outlined a new framework for preparing future engineers. He highlighted that the value of software engineers in the AI era extends far beyond mere coding speed. Instead, the focus shifts to critical thinking, robust system design, ethical leadership, and creative collaboration with AI systems.
According to Fu, one key reason for the continued necessity of human engineers is their ability to understand AI deeply – not just how to prompt it, but ‘what’s under the hood,’ including models, data pipelines, limitations, and inherent risks. This comprehensive understanding is crucial as AI becomes increasingly integrated into every product and system we build. Furthermore, engineers are vital for leveraging AI to develop ‘production-ready software that is scalable, reliable with long-term maintainability,’ moving beyond simple prototypes that AI can generate.
While AI systems excel at generating thousands of lines of code, translating programming languages, designing user interfaces, and even creating project demos from natural language prompts, they possess significant limitations. AI struggles with comprehending the ‘whys’ behind actions, lacks understanding of business sustainability, can make weak decisions, and is prone to ‘hallucinations.’ Fu likened AI to a ‘very smart young programmer: very quick, intelligent, and requiring a person to give direction regarding vision, morals, and outcomes.’
The role of software engineers is evolving from being solely ‘code writers’ to becoming ‘visionaries, bridge-builders, and leaders.’ They are tasked with tackling ‘messy, ambiguous problems’ and guiding machines toward structured and meaningful outcomes. This includes acting as system architects who design optimal solutions and ethical technologists who ensure that the solutions developed truly benefit human beings.
Also Read:
- Mark Zuckerberg Foresees AI Writing Majority of Meta’s Code Within 18 Months
- Geoffrey Hinton Issues Stark Warning: AI May Develop Unintelligible Languages, Reshaping Future Careers
For aspiring and current software engineers, Fu stressed several crucial skills for thriving in this AI-enhanced world. These include practicing strong communication and collaboration, embracing AI as a creative partner rather than fearing it, and delving into advanced AI concepts such as Large Language Models (LLMs), generative AI, model fine-tuning, and Retrieval Augmented Generation (RAG). He advised engineers to discuss projects with AI and delegate tasks to it as if it were a teammate. Above all, Fu emphasized the importance of adaptability, stating, ‘tools change principles last so you should always focus on learning how to learn.’ In an era where everyone might be able to code a little, those who master the craft and lead the path will be the true leaders.
