Unverified50% confidenceOpinionExact time
In AI application development, data preparation constitutes the majority of the workload, while writing model invocation code often accounts for only a small portion.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverified小到中等规模的编程任务（想法到代码）已经基本被AI解决，但系统设计和复杂工程架构仍是AI短板80% similarUnverifiedIn a typical AI feature deployment, the model calls themselves may account for only 5%–10% of the entire system's codebase, with the remaining 90% or more being engineering infrastructure built around the model.79% similarUnverifiedAI tends toward minimum viable delivery in code generation; when requirement descriptions lack sufficient information, the model tends to choose the most conservative and certain implementation path rather than inferring the user's implicit needs on its own.78% similarUnverifiedAI programming workflows should involve sufficient upfront planning before writing code, including defining project requirements and functional boundaries78% similarUnverifiedAI coding models need to process large amounts of context when completing tasks, including the current file, relevant codebases, project structure, and dependencies, which constitute a massive volume of input tokens.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/905015API
curl https://kongchang.com/api/v1/knowledge/claims/905015MCP
get_claim(id=905015)