17:32
AI Expert WARNING: Computer Science is DEAD, Study THIS Instead
Sajjaad Khader
Overview
This video challenges the traditional path of studying computer science and learning to code, arguing that AI is rapidly automating many of the tasks previously performed by junior software engineers. The CEO of an AI startup, Marcus Low, explains that while AI can now handle 95% of what engineers used to do, the role is evolving. Instead of building individual components, engineers will focus on building and managing the 'factories' that produce software. The video highlights the shift away from rote memorization and LeetCode-style interviews towards problem-solving with AI tools, emphasizing skills like understanding human needs, system design, and adaptability in a rapidly changing technological landscape.
How was this?
Save this permanently with flashcards, quizzes, and AI chat
Chapters
- AI can now perform approximately 95% of tasks previously done by software engineers.
- The definition of a software engineer is shifting from building individual product components to building the systems ('factories') that produce software.
- Large tech companies are hiring fewer junior engineers, partly due to AI automation, but early-stage AI-focused companies are hiring significantly.
Understanding this fundamental shift is crucial for aspiring and current software engineers to adapt their skills and career focus to remain relevant.
The industrial revolution saw engineers move from direct production to designing manufacturing processes; software engineering is undergoing a similar transition.
- LeetCode-style interviews are becoming obsolete because AI can easily solve these problems.
- Companies are moving towards unstudied, first-principles problems that require abstract thinking and problem-solving.
- Interviews now often require candidates to use AI tools, and some problems are designed to be nearly impossible without AI assistance.
This signals a move away from memorization-based skills towards practical application and AI-assisted problem-solving, changing how technical talent is assessed.
An interview question involves creating an API boundary that takes code as input, extracting user-editable strings, and storing them, a task made feasible with AI but difficult without it.
- Engineers need to become proficient in using AI tools to build products end-to-end.
- A key strategy is to 'build your own replacement' by creating tools or agents that automate your current tasks.
- Understanding human behavior, needs, and preferences is becoming increasingly valuable as AI handles technical execution.
Developing these future-proof skills will enable engineers to leverage AI rather than be replaced by it, creating new opportunities and higher impact.
An AI native engineer, Arnoff, impressed his CEO by identifying a niche problem (mainframe to Postgress conversion), building a solution end-to-end using AI, and shipping it to customers.
- Learning to code for a professional career is discouraged; focus should be on higher-level problem-solving and system design.
- Studying humanities (like history, literature) provides essential skills in understanding human patterns, motivations, and changing preferences.
- The ability to adapt to rapid change and embrace new technologies is the most critical skill for long-term career success.
This perspective suggests a shift in educational focus from purely technical skills to a blend of technical proficiency with deep human understanding and continuous learning.
The rapid shift in user perception of self-driving cars (from 'incredible' to 'boring') illustrates the fickle and rapidly changing nature of human preferences that AI struggles to predict.
- Anything.com is a platform enabling non-technical users to build end-to-end products using AI agents.
- The company operates a 'software engineering factory' that continuously monitors, prioritizes, and proposes code changes, automating much of the development process.
- Engineers at Anything.com leverage AI extensively, with unlimited tokens, and focus on building complex systems and improving the AI factory's capabilities.
This demonstrates a practical, real-world application of the AI-driven software development paradigm discussed, showcasing new roles and opportunities.
A feature allowing users to build apps by texting a number was primarily developed by the AI factory, with human engineers providing feedback and steering the process.
Key takeaways
- The role of a software engineer is evolving from writing code to orchestrating and managing AI systems that write code.
- Traditional coding skills and rote memorization are becoming less valuable compared to problem-solving, system design, and AI tool proficiency.
- Aspiring engineers should focus on building products with AI, understanding human needs, and developing adaptability rather than solely on learning to code.
- The future of software development involves 'factories' or systems that automate code generation, testing, and deployment.
- Skills derived from humanities, such as understanding human behavior and historical patterns, are becoming increasingly important in a tech-dominated world.
- Embrace rapid technological change and learn to leverage AI tools to build your own 'replacement' systems, thereby increasing your value.
- Early-stage AI companies are actively hiring, seeking individuals who can navigate and build within the new 'agentic economy'.
Key terms
AI AutomationSoftware Engineering EvolutionAgentic EconomySystem OrchestrationFirst Principles ThinkingHumanitiesAI NativeSoftware Engineering FactoryPrompt EngineeringEnd-to-End Product Development
Test your understanding
- How has the definition of a software engineer's role changed due to AI, and what is the new focus?
- Why are LeetCode-style interviews becoming obsolete, and what types of interview questions are replacing them?
- What does the advice 'build your own replacement' mean in the context of AI and software engineering?
- Why might studying humanities be considered more valuable than traditional computer science for a future career in tech?
- How does a 'software engineering factory' operate, and what is its significance in the development process?