The Engineering Dilemma in Enterprise AI Adoption
yangyi · x · 2026-07-18
The author points out a subtle contradiction in enterprise AI adoption in China: if a team is purely an engineering unit without relying on training, customer acquisition, or dispatching to drive implementation, they must answer a practical question—within short life-cycles, how many copies can you sell of the capabilities you initially developed at a loss?
The author notes that many projects invest heavily in building knowledge bases and workflows for their first few clients, earning one-off consulting fees without forming true infrastructure. As model capabilities improve, the bottlenecks they originally solved are quickly flattened by the models themselves, leaving only half a mile of the last mile.
Therefore, the real value lies not in broadly solving enterprise problems, but in identifying segments across the entire pipeline that are highly repetitive, manageable by models, but still rely on industry-specific engineering. The goal is to quickly solidify these into long-term assets and competitive moats during this cycle of continuous model advancement.
More from Companies & People
- A post says AI teams should drop the research scientist vs engineer split — jsuarez · 2026-07-22
- YC startup Vendo launches an open-source customization layer for user-built micro-apps — ycombinator · 2026-07-22
- Prescience launches publicly as an AI-native health insurance company — ycombinator · 2026-07-22
- SF AI crowd swaps poker for a bullet and bughouse chess night — Jackyhuang · 2026-07-22
- Polymarket puts Anthropic’s year-end IPO odds at 64% amid patent suit — Polymarket · 2026-07-22
- University of Tennessee sues Anthropic over machine learning and neuromorphic patents — Polymarket · 2026-07-22