Organizations: Chalmers University of Technology Gothenburg, Sweden · University College London London, United Kingdom · Volvo Group Gothenburg, Sweden
Multidisciplinary Software Development (MSD) requires domain experts and developers to collaborate across incompatible formalisms and separate artifact sets. Today, even with AI coding assistants like GitHub Copilot, this process remains inefficient; individual coding tasks are semi-automated, but the workflow connecting domain knowledge to implementation is not. Developers and experts still lack a shared view, resulting in repeated coordination, clarification rounds, and error-prone handoffs. We address this gap through a graph-based workflow optimization approach that progressively replaces manual coordination with LLM-powered services, enabling incremental adoption without disrupting established practices. We evaluate our approach on spapi, a production in-vehicle API system at Volvo Group involving 192 endpoints, 420 properties, and 776 CAN signals across six functional domains. The automated workflow achieves 93.7% F1 score while reducing per-API development time from approximately 5 hours to under 7 minutes, saving an estimated 979 engineering hours. In production, the system received high satisfaction from both domain experts and developers, with all participants reporting full satisfaction with communication efficiency.
Figures & tables
Figure 1. Comparison of workflows in the Multidisciplinary Software Development (MSD) process: (a) a typical MSD workflow and (b) optimized workflow through automated translation via our graph-based approach.
Figure 2. Iterative workflow optimization framework for vehicle API generation. The upper part illustrates graph-level optimization, depicting how the initial manual workflow undergoes three iterative refinements to achieve a highly automated workflow. The lower-right part illustrates the final deployed workflow, which comprises three server nodes. The lower-left part illustrates the detailed structure of each node.
Dimension
Score
Description
Complexity
-1
Task or exchange removed or combined
0
No change
+1
Task or exchange added
Automation
-1
Task becomes more manual
0
No change
+1
Task becomes more automatic
Table 1. Scoring scheme for transformation impact across three dimensions
Figure 3. A simplified DSPy signature for Signal R/W Synthesis. The LLM generates Python functions to read or write CAN signals based on signal metadata.
Figure 4. Representative prompt used in the Signal-Property matching task (simplified for clarity).
Figure 5. Example of a FastAPI endpoint boilerplate template.
Figure 6. A representative prompt used in the API Endpoint synthesis task (simplified for clarity).
Iteration
Comp.
Auto.
Comm.
1: Signal R/W
-1
+1
+2
2: Signal-Property
-2
+1
+2
3: Property-Endpoint
-2
+1
+2
Net change
-5
+3
+6
Table 2. Cumulative transformation impact across three iterations.
Large Language Models (LLMs) have accelerated the adoption of software development agents, now widely available as Integrated Development Environment (IDE) extensions and standalone applications. While these agents are typically general-purpose, it remains unclear whether specialist agents justify their additional development effort. We investigate this question in the context of business process automation, focusing on the transformation of Business Process Model and Notation (BPMN) diagrams into executable agentic workflows. Since BPMN specifies explicit control-flow semantics, we focus on deterministic workflows in which a fixed process model and inputs uniquely determine the executed path. We introduce a specialist workflow for this task and compare it against generalist agents such as Roo and Cline. Our results show that the specialist solution produces agents that outperform generalist baselines by approximately 9-20 percentage points in tool-use exactness, 2-4x in penalty-adjusted latency, and 3x fewer tool-call errors, while reducing generation token cost by over 95% and eliminating repair iterations. We also find that generalist agents generate code inconsistently in both functionality and quality, limiting their suitability for industrial settings where reliability and maintainability are essential.
Harris Borman, Herman Wandabwa, Fusun Yu +4
Commonwealth Bank of Australia, Sydney, Australia.
This paper explores the potential of Artificial Intelligence (AI) to enable end-to-end automotive design and software development for Software-Defined Vehicles (SDVs). Based on our contributions within the Central Car Server (CeCaS) and follow-up EU-funded projects, we present an AI-driven toolchain integrating Large Language Models (LLMs) with Model-Driven Engineering (MDE) to automate the transformation of natural-language requirements into system architectures, executable software, and validation artifacts. The approach emphasizes traceability, consistency, and compliance with safety-critical automotive development requirements. Validation is performed on a centralized vehicle computing platform supporting adaptive cruise control and emergency braking, tested in Hardware-in-the-Loop (HiL) and Software-in-the-Loop (SiL) environments. We discuss current capabilities, practical limitations, and future directions toward increasingly automated, reliable, and quality-assured automotive software development with minimal human intervention.
Interoperability between heterogeneous modeling tools remains a significant challenge in Model-Driven Engineering (MDE), particularly in the automotive domain where multiple modeling languages, as well as defacto standard proprietary and open-source tools coexist. This paper presents an LLM-driven approach for automated model interoperability by considering two relevant aspects: 1) mapping model instances to a target metamodel 2) merging of metamodels. The proposed methodology is demonstrated through transformations involving Ecore and SysML v2 based metamodels and incorporates structural validation of generated model instances against user-defined target models. Automotive case studies illustrate the feasibility of the approach and show that large language models can significantly reduce manual transformation effort while generating structurally valid target models for cross-tool interoperability.
Nenad Petrovic, Jiajie Zhang, Vahid Zolfaghari +1
Chair of Robotics, Artificial Intelligence and Real-Time Systems Technical University of Munich, Munich, Germany