Amazon and SimScale report measurable value from AI in engineering

Recent industry findings highlight the growing impact of artificial intelligence on engineering productivity and development cycles. A survey conducted by SimScale involving 300 engineers found that 92% of respondents see measurable value from AI applications. Key benefits reported include faster engineering and development cycles (46%), increased productivity (39%), and reduced simulation or testing efforts (23%). For these professionals, saving engineering hours is a more significant metric than cost reduction.

In a separate internal memo, Terk Terkowitz, Vice President and Distinguished Engineer at Worldwide Amazon Stores, advised engineers to focus on long-term outcomes rather than input-based metrics like token usage or lines of code. Terkowitz cautioned that prioritizing metrics such as ‘tokenmaxxing’ can lead to increased costs and complexity, which contradicts core principles of frugality and simplicity. He emphasized that the true measure of AI success is its ability to drive productivity and delivery rather than surface-level tool adoption.