Dumping Cursor for VSCode + Claude Code
A developer explains why they are switching from the AI-native Cursor editor to using Claude Code within VSCode for AI-powered programming.
Daniel Miessler is a cybersecurity and AI engineer turned founder, based in the San Francisco Bay Area. He shares insights on cybersecurity, artificial intelligence, technology, and human behavior through essays, tutorials, and technical content on his blog.
91 articles from this blog
A developer explains why they are switching from the AI-native Cursor editor to using Claude Code within VSCode for AI-powered programming.
An Apple fan reflects on Google I/O 2025, expressing fear for Apple's future as Google pivots to an AI-first vision with Gemini.
Argues that the 'LLMs only predict tokens' critique fails, as human brains are similarly opaque systems that process input to produce output without conscious understanding.
AI will widen educational gaps by default, making great students extraordinary and struggling students worse, unless systems are designed to close them.
A cybersecurity expert contrasts two groups of colleagues: AI skeptics who see it as overhyped and harmful, and those who recognize its transformative potential.
Argues that AI security levels are determined by market forces and user behavior, not by individual efforts, and will reach a functional equilibrium.
Explores how AI could transform large tech companies into vast networks of micro-startups, shifting the role of developers to broader builders.
Explores a human-centric definition of ASI and proposes a scalable, iterative methodology for achieving both AGI and ASI.
Argues that coding is fundamental thinking and creativity, and that tech professionals must adapt beyond just execution to avoid career risk in the AI era.
Explores AI's potential as a universal problem-solving tool, framing any challenge as a transition from a current state to a desired state.
Argues that current AI models are already capable of achieving Functional AGI through better orchestration of existing systems, not new model breakthroughs.
Distinguishes between Functional AGI (replacing knowledge workers) and Technical AGI (true generalization), arguing Functional AGI's societal impact matters most.
A guide outlining the top 10 personal and business risks associated with AI adoption, including failure to adopt and security concerns.
Explores how AI will soon enable rapid novel creation through prompting and verbal narration, transforming the writing process.
Apple plans to transform Siri into a powerful AI by integrating it deeply with the Apple ecosystem, including personal context and cross-app features.
Analyzes the massive total addressable market for AI, arguing it encompasses the cost of human labor and new business value.
Argues for a balanced approach to AI, using it for discovery but doing important learning and thinking tasks manually to preserve meaning and growth.
Learn powerful Neovim/Vim commands for precisely copying and moving lines of code to boost your editing efficiency.
AI will augment founders, enabling them to perform vast amounts of work without employees, fundamentally reshaping the economy and employment.
Explores a method using a 'Judging AI' (like o1-preview) to evaluate the performance of other AI models on tasks, relative to human capability.