2 min lesson
Three differentiators sharper than the scorecard
Give your one-line thesis on why Cursor wins and where it is most exposed.
Step 1 of 2
Three differentiators sharper than the scorecard
The generic scorecard above wins the “name the field” round. These three earn the “you actually understand the product” round - each is a defensible take an interviewer can't get from the website.
Cursor runs Claude, OpenAI/Codex, Gemini, Grok, open-source or its own ComposerCursor's own fast coding model, tuned for the editor and priced well below frontier models; the recommended day-to-day model for executing a plan. Press Enter for the full definition. - and you switch the moment a new SOTA model ships, roughly monthly.
Lock yourself to Claude Code and you effectively only have Claude. Optionality is structural, not a feature.
Cursor isn't a desktop app piping prompts to providers. The proprietary cloud harness does optimization, context enrichment, caching and dynamic context discoveryThe agent pulling only the relevant parts of files, tools and MCP servers into context as needed, instead of loading everything up front. Press Enter for the full definition..
It cuts agent tokens ~47% even on other vendors' models, and the same model performs better here than in Claude Code because the harness is tuned per model.
Bring-your-own-key is supported (AWS Bedrock), but it's a degraded experience - going through Cursor directly is faster and yields better results.
The harness optimizations are exactly what BYOK skips. Naming this concedes a real option while steering off it.
Cursor wins because it made codebase-aware multi-file editing feel fast and trustworthy enough to live in all day and that's a product-and-retrieval moat, not a model moat. The under-discussed edge is the harness - it's not a thin proxy; it reduces agent tokens about 47% even on other vendors' models, so the same model performs better in Cursor than in a single-vendor tool. Where it's most exposed: if advanced models commoditize and an incumbent with GitHub-grade distribution closes the experience gap, the wedge narrows. So my job is to widen the habit and the team lock-in faster than that gap closes.