AST

What is Vibe Coding? Why AI Code Review is the Future In February 2025, Andrej Karpathy, former Tesla AI Lead and OpenAI co-founder, tweeted a phrase that would define a new paradigm in software development: “There’s a new kind of coding I call ‘vibe coding’, where you fully give in to the vibes, embrace exponentials, and forget that the code even exists.” In the months since, vibe coding has evolved from a catchy meme into one of the most significant AI software engineering trends of the decade. Non-technical founders are shipping complex web apps in hours, product managers are replacing spreadsheet systems with automated dashboards, and engineers are building features at a velocity that was previously unimaginable. ...

Prerequisite: Familiarity with the concepts introduced in Part 1 — Vibe Coding Non Technical. Review it first if the terminology in this part is unfamiliar. Context Engineering for Codebase AI Code Reviewers Answer-first: Context engineering for codebase AI code review extracts AST function signatures, repository rules, and model dependencies to build token-budgeted prompt contexts, reducing LLM reviewer false positives from 42% to under 4%. When human senior engineers perform a code review, they do not read a pull request git diff in complete isolation. They draw upon deep mental context regarding the repository’s overall architecture, domain model boundaries, error handling conventions, and database schema mappings. ...