Context engineering is quickly becoming the backbone of modern artificial intelligence as organizations push for systems that stay accurate, grounded and predictable while handling increasingly complex tasks.
Enterprises now rely on AI to power high-speed decisions and agentic workflows, raising expectations for precision and providing the guardrails that help keep automation trustworthy. This shift has created an urgent need for approaches that help models operate on the right data, in the right boundaries, at the right time, according to Ken Exner (pictured), chief product officer of Elasticsearch B.V.
“I think context engineering, it’s one of those very popular terms these days,” Exner said. “I think it’s because people are starting to build AI applications, agentic AI applications. What they’re realizing is that the most important part of building this agentic AI application is making sure that the LLMs have the right data, have the right data to ground them on the right context, to scope the actions of that agent.”
Exner spoke with Jackie McGuire as part theCUBE’s coverage of AWS re:Invent. During an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio, they explored Elastic’s perspective on context engineering, agentic AI, observability and the evolving standards that shape reliable AI systems. (* Disclosure below.)
How context engineering is shaping AI’s next phase
The deeper enterprises move into agentic automation, the more they encounter the limits of traditional prompt engineering. Teams are discovering that successful AI systems depend not just on models, but on the continuous delivery of context — grounding data, personalization cues and domain knowledge that guide how an AI interprets each request. This evolution is elevating context engineering into a central pillar of AI development, Exner pointed out.
“This entire field of getting the right data to an LLM is increasingly being called context engineering,” he said. “I personally think it’s the most important thing in building AI applications that work, that succeed, that do the right things. I think you’re going to hear this term, context engineering, a lot over the next year because context engineering is what is vital to do an AI right.”
The need becomes even more pressing as agentic systems grow more autonomous. When models lack the context they need, they fall back on general knowledge, leading to unpredictable or inaccurate behavior. Developers are embracing techniques such as retrieval, tool-based reasoning and memory systems to ensure models stay aligned with the data that matters most, Exner emphasized.
“An LLM is essentially a predictive system. It’s predicting the next token, and it’s doing this by a process of reasoning about what is the best answer for the next token,” he added. “In order to do that, it has to be scoped to the right data. If you don’t scope something, it’s going to use its general foundational knowledge to provide an answer. If you want to make sure that it’s scoped to a particular corpus of data or that it understands who you are personally … you need to make sure it has that context.”
Building smarter agents with streamlined tools and stronger oversight
Platforms such as Elastic Agent Builder help operationalize these practices by pairing data indexing, connectors and customizable prompts with easy pathways to create agents tailored to specific workloads. This streamlines how teams build applications grounded in the data that drives their business, accelerating development and improving reliability out of the gate.
“We actually provide you an out-of-the-box conversational agent by default for any index in Elastic, which means that you can start having a conversation with your data,” Exner said. “It is a very easy way to start having an agentic application on top of any data, any custom data, because we give you one out of the box by default and then allow you to customize it.”
Evaluation and observability are also becoming essential checks within context engineering. Enterprises increasingly treat agents as first-class systems that need performance validation, relevance testing and the same level of oversight used for mission-critical infrastructure. That rigor is shaping a new foundation for AI operations, according to Exner.
“A critical part of doing context engineering is evaluation,” he said. “It is the observability and evaluation to make sure that you’re producing the right results. Evals and context engineering are kind of like unit tests. They help you test the quality and efficacy of what you’re doing. You also have the equivalent of an integration test, which is like LLM as judge, which is trying to assess the entire answer and trying to make sure that you have the most relevant, the most accurate information.”
Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of AWS re:Invent:
(* Disclosure: Elastic sponsored this segment of theCUBE. Neither Elastic nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)
Photo: SiliconANGLE
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
- 15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more
- 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.
About SiliconANGLE Media
SiliconANGLE Media is a recognized leader in digital media innovation, uniting breakthrough technology, strategic insights and real-time audience engagement. As the parent company of SiliconANGLE, theCUBE Network, theCUBE Research, CUBE365, theCUBE AI and theCUBE SuperStudios — with flagship locations in Silicon Valley and the New York Stock Exchange — SiliconANGLE Media operates at the intersection of media, technology and AI.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.
