About

Dr. Kosa Goucher-Lambert is an Associate Professor of Mechanical Engineering at the University of California, Berkeley, and an Affiliate Faculty member in the Jacobs Institute of Design Innovation, Berkeley Institute of Design, and UC Berkeley Human Computer Interaction Group. He serves as the Associate Director of the UC Berkeley Master of Design program. His expertise lies in engineering design theory, methods, and automation, with research focusing on decision-making applied to engineering teams and individuals, ideation and creativity, analogical reasoning in design, preference modeling, and design attribute optimization. His work also encompasses design cognition, neuroimaging methods applied to design, sustainable design, new product development, crowdsourcing, and collaboration. Dr. Goucher-Lambert has received several awards, including an NSF CAREER Award, the 2022 ASME Design Theory and Methodology Young Investigator Award, and the 2019 Excellence in Design Science Award. He has also earned multiple best paper awards from the American Society of Mechanical Engineers and the Design Society.

Research topics

• Computer Science
• Artificial Intelligence
• Psychology
• Engineering drawing
• Knowledge management
• Engineering
• Geometry
• Cognitive psychology
• Mathematics
• Social psychology
• Marine engineering

Selected publications

• Developing Design Theory Using Large Language Model Agents: A Case Study of C-K Theory

  Journal of Mechanical Design · 2026-02-26

  articleSenior author

  Abstract Computational simulations have long been used for design and engineering analysis, but there is also a rich history of using simulations to develop design theory. The present work draws inspiration from design, management, psychology, and other fields that have used simulations to develop theory and integrates recent advances in large language models (LLMs). While design theory has been used to improve the development of generative design tools, this study goes the other direction and uses generative AI-based simulations to further develop design theory. We take C-K theory as a case study to demonstrate this approach. The simulations used computational agents fueled by large language models designed to adhere to the C-K theory methodology in both wording and framework. These were evaluated using a mixed-methods approach across three studies, including two simulation experiments and one conventional content analysis of qualitative, LLM-generated, C-K transition rationale statements. The results reveal that concept-to-concept transitions were the predominant C-K operation and that concept diversity trended downward across various experimental conditions. These results are in contrast to the generic generativity described in human-based C-K theory and highlight gaps between C-K as a design theory and C-K in a generative AI-based computational simulation form, suggesting future directions for operational and theoretical development. Ultimately, this case study demonstrates that design theories can be simulated computationally using LLM agents, albeit with differences and potential limits when compared to humans, thus providing the design research community with a new approach to further developing design theories.

  PublisherDOI

• Characterizing Design Rationales Using Computational Linguistics and Human Evaluations

  Journal of Mechanical Design · 2025-07-25

  articleSenior author

  Abstract Design rationales capture the explicit justifications behind design decisions. Often, rationales vary in the content and depth of information, making the study and comparison of rationales challenging. This project aims to characterize design rationales and develop a standardized approach to assess design rationale quality at scale. In total, 2250 rationales were machine-generated across two different representations and evaluated by two raters across five dimensions of quality. Rationales were then characterized using natural language processing techniques, resulting in 108 linguistic features for each rationale. The linguistic features were used to build predictive models for each quality dimension. The main results identify correlations between linguistic features and human evaluations, show that structured rationales were rated higher than unstructured rationales across the five dimensions, and present a model to predict rationale quality for new texts. These findings help inform strategies to improve human- and machine-generated rationales. The predictive tool offers a scalable method for evaluating different representations of rationale, thereby supporting more effective documentation practices.

  PublisherDOI

• Generating preinventive structures: AI-driven creativity in product repurposing

  Proceedings of the Design Society · 2025-08-01

  articleOpen accessSenior author

  ABSTRACT: This study presents an AI-driven method for generating preinventive structures - initial precursors to creative design concepts - using the Geneplore model as a theoretical framework. Multimodal AI is leveraged to derive preinventive structures from combinations of components of an existing product. This method is evaluated by comparing AI-generated structures of a product to those reverse identified from real repurposing solutions for the same product (IKEA hacks). The appearance of AI-generated preinventive structures in the repurposed designs suggests that this method can inspire and lead to viable design concepts. Implications extend to sustainable design, creative ideation, and the theory-driven development of design methods that support design in constrained solution spaces. Future work can refine these approaches and investigate broader applications in diverse design contexts.

  PublisherOA PDFDOI

• The Future of Design: Five Key Snapshots

  Journal of Mechanical Design · 2025-11-18

  articleOpen access1st authorCorresponding

  Abstract This editorial is the first in a series of editorials exploring the future of engineering design from the perspective of thought-leaders within the engineering design community. Five early- and mid-career researchers were invited to present their perspectives at a meeting of the Design Society in March 2025 on the campus of the Georgia Institute of Technology. Jessica Menold focused on teaming and collaboration in the context of artificial intelligence (AI); Kosa Goucher-Lambert on design cognition; Astrid Layton on sustainable and resilient design; Mohsen Moghaddam on virtual reality (VR)/augmented reality (AR)/extended reality (XR) in engineering design; and Zhenghui Sha on the design of complex sociotechnical systems. Each presentation was followed by roundtable discussions of the challenges and opportunities posed by the speaker. The presentations and audience discussions are summarized in this editorial, along with a brief overview of some of the opportunities for further work. The goal of this editorial series is to inform the broader community of the progress, challenges, and opportunities associated with important themes within our research community and to offer a starting point for those who seek to investigate these topics and continue to advance the state-of-the-art in our field.

  PublisherOA PDFDOI

• Simulating Design Theory Using LLM Agents: A Case Study of C-K Theory

  2025-08-17 · 1 citations

  articleSenior author

  Abstract In the majority of computational simulations developed for engineering design research, the focus is on simulation for the purpose of analysis, such as simulating stresses to identify yield or fracture points in structures. However, what about simulating design theory for the purpose of developing theory? In fields like organizational psychology, simulations have proven valuable in predicting behaviors and understanding decision-making processes. This work draws inspiration from those fields to investigate its applicability in design theory. This work use C-K design theory as a representative case study to demonstrate this approach. We designed a simulation using computational agents fueled by large language models. The simulation was designed to adhere to the C-K theory methodology in both wording and framework. The results of the simulation were evaluated utilizing a mix of both qualitative and quantitative methods. Findings from the results reveal that the concept to concept transition was the predominant operation and that diversity trended downwards across multiple experimental conditions. These findings from the simulation highlight gaps in C-K theory and suggest directions for future theoretical development. Ultimately, this case study demonstrates that design theories can be effectively simulated computationally, enabling the design research community to better understand and develop improved design theories.

  PublisherDOI

• Human-Gen AI Co-Design: Exploring Factors Impacting Trust Calibration

  2025-08-17 · 1 citations

  articleOpen access

  Abstract The process of generating ideas during co-design with a Generative AI (GenAI) system requires the gradual calibration of trust in that system. Trust plays a pivotal role in shaping human interactions with technology, and developing well-calibrated trust is essential for the effective use and integration of GenAI. Proper trust calibration helps prevent underutilization of the system’s capabilities and dissatisfaction with its output. For engineers and system designers, trust is particularly important as it directly influences user responses, system adoption, and overall engagement with new technologies. To explore the factors that influence trust fluctuation when co-designing with a GenAI system, we analyzed 12 hours of conceptual human-AI co-design sessions using a custom GenAI system capable of producing images across various generation modes from convergent-divergent to abstract-concrete, and combining text and sketch prompting. Focusing on each moment of interaction with GenAI-generated images, we conducted an incremental and qualitative coding of each trust-related extract from think-aloud protocols. Through this approach, we identified 23 key factors that cause fluctuations in trust. Our findings reveal a complex network of factors that impact trust calibration, offering insights into how GenAI systems can be designed to facilitate faster and more effective trust-building in human-GenAI collaborations.

  PublisherOA PDFDOI

• RECALL-MM: A Multimodal Dataset of Consumer Product Recalls for Risk Analysis using Computational Methods and Large Language Models

  ArXiv.org · 2025-03-29

  preprintOpen accessSenior author

  Product recalls provide valuable insights into potential risks and hazards within the engineering design process, yet their full potential remains underutilized. In this study, we curate data from the United States Consumer Product Safety Commission (CPSC) recalls database to develop a multimodal dataset, RECALL-MM, that informs data-driven risk assessment using historical information, and augment it using generative methods. Patterns in the dataset highlight specific areas where improved safety measures could have significant impact. We extend our analysis by demonstrating interactive clustering maps that embed all recalls into a shared latent space based on recall descriptions and product names. Leveraging these data-driven tools, we explore three case studies to demonstrate the dataset's utility in identifying product risks and guiding safer design decisions. The first two case studies illustrate how designers can visualize patterns across recalled products and situate new product ideas within the broader recall landscape to proactively anticipate hazards. In the third case study, we extend our approach by employing a large language model (LLM) to predict potential hazards based solely on product images. This demonstrates the model's ability to leverage visual context to identify risk factors, revealing strong alignment with historical recall data across many hazard categories. However, the analysis also highlights areas where hazard prediction remains challenging, underscoring the importance of risk awareness throughout the design process. Collectively, this work aims to bridge the gap between historical recall data and future product safety, presenting a scalable, data-driven approach to safer engineering design.

  PublisherOA PDFDOI

• RECALL-MM: A Multimodal Dataset of Consumer Product Recalls for Risk Analysis Using Computational Methods and Large Language Models

  2025-08-17

  articleSenior author

  Abstract Product recalls provide valuable insights into potential risks and hazards within the engineering design process, yet their full potential remains underutilized. In this study, we curate data from the United States Consumer Product Safety Commission (CPSC) recalls database to develop a multimodal dataset, RECALL-MM, that informs data-driven risk assessment using historical information, and augment it using generative methods. Patterns in the dataset highlight specific areas where improved safety measures could have significant impact. We extend our analysis by demonstrating interactive clustering maps that embed all recalls into a shared latent space based on recall descriptions and product names. Leveraging these data-driven tools, we explore three case studies to demonstrate the dataset’s utility in identifying product risks and guiding safer design decisions. The first two case studies illustrate how designers can visualize patterns across recalled products and situate new product ideas within the broader recall landscape to proactively anticipate hazards. In the third case study, we extend our approach by employing a large language model (LLM) to predict potential hazards based solely on product images. This demonstrates the model’s ability to leverage visual context to identify risk factors, revealing strong alignment with historical recall data across many hazard categories. However, the analysis also highlights areas where hazard prediction remains challenging, underscoring the importance of risk awareness throughout the design process. Collectively, this work aims to bridge the gap between historical recall data and future product safety, presenting a scalable, data-driven approach to safer engineering design.

  PublisherDOI

• Empowering designers to create life cycle informed products: heuristics for extracting insights from LCA reports

  Design Science · 2025-01-01 · 2 citations

  articleOpen accessSenior authorCorresponding

  Abstract Life cycle assessment (LCA) reports are commonly used for sustainability documentation, but extracting useful information from them is challenging and requires expert oversight. Designers frequently face technical obstacles and time constraints when interpreting LCA documents. As AI-driven tools become increasingly integrated into design workflows, there is an opportunity to improve access to sustainability data. This study used a mixed-methods approach to develop life cycle design heuristics to help non-LCA experts acquire relevant design knowledge from LCA reports. Developed through in-depth interviews with LCA experts ( n = 9), these heuristics revealed five prominent categories of information: (1) scope of analysis, (2) priority components, (3) eco hotspots, (4) key metrics, and (5) design strategies. The utility of these heuristics was tested in a need-finding study with designers ( n = 17), who annotated an LCA report using the heuristics. Findings suggest a need for additional support to help designers contextualize quantitative metrics (e.g., carbon footprints) and suggest relevant design strategies. A follow-up reflective interview study with LCA experts gathered feedback on the heuristics. These heuristics offer designers a framework for engaging with sustainability data, supporting product redesign, and a foundation for AI-assisted knowledge extraction to integrate life cycle information into design workflows efficiently.

  PublisherOA PDFDOI

• Extreme Design: An Editorial on a New Research Framework Within Engineering Design

  Journal of Mechanical Design · 2025-09-19

  article

  Abstract Extreme design (XD) is a proposed research framework addressing engineering design's outer edges of complexity and uncertainty. As the scale and urgency of global challenges grow, such as climate change, autonomous systems, and aging populations, so does the need for design approaches that go beyond conventional methods and models. XD offers a way to approach design problems that are dynamic, interdisciplinary, and fundamentally hard to frame but have humanity at their core. This editorial introduces XD as a framework for developing new theories, methods, and tools suited to extreme conditions. It outlines research opportunities in adaptive systems, creative processes, multiscale prototyping, convergent collaboration, and sustainability. A complexity–uncertainty matrix positions XD relative to conventional and emerging design approaches. We aim to open a conversation—not to define XD fully, but to signal its necessity and invite the design research community to explore and shape it. The challenges ahead will not be solved by incremental improvement in how we approach design. They will require something new. We feel XD is a step in that direction.

  PublisherDOI

Recent grants

• CAREER: Synthesizing Neuroimaging and Data Driven Approaches to Discover Inspirational Examples in Design

  NSF · $581k · 2022–2027

Frequent coauthors

• Jonathan Cagan

  Carnegie Mellon University

  25 shared

• Vivek Rao

  18 shared

• Alice M. Agogino

  16 shared

• Elisa Kwon

  10 shared

• Ananya Nandy

  Vanderbilt University Medical Center

  9 shared

• Christopher McComb

  9 shared

• Kenneth Kotovsky

  Carnegie Mellon University

  8 shared

• Yakira Mirabito

  University of California, Berkeley

  8 shared

Education

• Ph.D., Mechanical Engineering

  University of California, Berkeley

  2015

• M.S., Mechanical Engineering

  University of California, Berkeley

  2012

• B.S., Mechanical Engineering

  University of California, Berkeley

  2010

Awards & honors

• NSF CAREER Award
• 2022 ASME Design Theory and Methodology Young Investigator A…
• 2019 Excellence in Design Science Award
• several best paper awards from the American Society of Mecha…