Mainstream AI technology is powerful but imperfect. Today’s organizations require reliable systems and more than off-the-shelf technology. The issues have led to a rapidly developing AI engineering (AIE ) practice with an ecosystem of consultants and technology providers. AI engineers help organizations to develop AI systems that can team with people to do sensemaking and reasoning.
- Sensemaking is collecting, organizing, summarizing, and extracting information. Sensemaking technology (like large language models) is a powerful component of most AI applications.
- Reasoning is modeling how the world works, planning, and taking actions. Reasoning is needed for scheduling, design, manufacturing, construction, logistics and other complex activities.
- Compound agentic systems enable a modular design approach enabling AI engineers make design and combine sensemaking and reasoning. This approach is increasingly used in creating state of the art AI systems.
This project is about state of the art practices for creating AI applications.
The paper below is intended for people and organizations who need an understanding of AI technology and the opportunities, limitations and risks. It explains the research breakthroughs, the state of the art, how AIE bridges the gap for creating reliable applications, and the whitespace and ideas for advancing the art. The goal is to demystify the research and engineering basis for creating powerful and reliable AI applications.
The “AIE” Paper: Addressing Sensemaking, Reasoning, and Explainability
- What is the state of the art in building AI applications on foundation models?
- What research advances are enabling AIE?
- How does AIE address sensemaking, reasoning, and explanation?
- What are the limitations and the whitespace for AIE?
Papers
Stefik, M., Gunning, D., Choi, J., Miller, T., Stumpf, S., Yang,, G.-Z. 2025. How AI Engineering Addresses Sensemaking, Reasoning, and Explainability. https://www.dropbox.com/scl/fi/76bv3kw0i73zifp37lgqo/2025-08-30-How-AI-Engineering-Addresses-Reasoning-and-Explainability.pdf?rlkey=hiax086p23whdlq4o5p4v2p20&dl=0