The new engine of the AI era... "Contextual Engineering" becomes the core of enterprise strategy

As enterprises begin to adopt “context engineering” as a core strategy to enhance the reliability and accuracy of AI systems, this technology is emerging as a new essential element in the AI era. At AWS re:Invent 2025, Elastic Chief Product Officer Ken Exner emphasized that large language models must be designed to operate “at the right time, with the right data, and within the appropriate scope” in order to deliver reliable results.

Exner stated, “Many enterprises today are facing the limitations of prompt engineering alone when implementing agent AI. To successfully build AI applications, it’s crucial to continuously provide the LLM with the correct context.” He referred to this as “context engineering” and predicted that it would become a core concept in future AI development.

As AI models gain increasing autonomy in decision-making and action, Exner pointed out the need to be vigilant about the risk of errors or uncertainty caused by missing context. To address this, supplementary approaches such as retrieval techniques, tool-based reasoning, and memory systems are being introduced. He explained, “LLMs are essentially systems that predict the next word, and only by performing this process within the proper data scope can we achieve more consistent and reliable results.”

To address these technical challenges, Elastic has developed the “Elastic Agent Builder” solution. This tool enables the creation of sophisticated agent applications by combining user-customized prompts with data indexing capabilities, and it includes built-in foundational conversational agents to help users easily create their own AI agents.

Standards for evaluating the success of context engineering are also being established. In this process, “evaluation” and “observability” play important roles. Exner stated, “We now need to treat agents like core systems and strengthen performance verification and quality testing. Such validation should operate like unit testing, while also incorporating integrated testing that uses LLMs as judges.”

Among enterprises seeking to ensure the reliability and sustainability of AI systems, context engineering is moving beyond a buzzword to become a substantive technological foundation and strategic approach. Looking ahead at the conference, Exner remarked, “In the coming year, we’ll hear the term ‘context engineering’ more frequently, and this field will play a decisive role in advancing AI to the next stage.”