sales engineering

This study addresses the limited understanding of how practitioners actually develop software engineering (SE) agents, particularly the lack of systematic investigation into the evolution of development workflows and core challenges. Through semi-structured interviews with 20 practitioners complemented by a survey of 80 respondents, this work proposes the first seven-stage workflow for SE agent development, revealing a paradigm shift toward “evaluation-driven iteration.” The research identifies that bottlenecks have moved beyond coding to non-coding tasks such as requirement specification, cross-role coordination, review, and deployment. It systematically characterizes six key challenges—including unreliable evaluation signals, accumulating comprehension debt, and behavioral drift induced by model updates—and synthesizes corresponding practical mitigation strategies.

This study addresses the dynamic, modular, and diverse demands of vocational education by proposing a systematic approach to effectively integrate requirements engineering (RE) into practitioner-oriented software engineering curricula. Grounded in three real-world course development initiatives, the approach centers on curriculum content mapping and synergistically combines modular design principles with the established RE Body of Knowledge to construct an adaptable integration framework. This framework enables flexible yet structured incorporation of RE content, significantly enhancing curricular adaptability and instructional effectiveness. Empirical validation through the implemented courses demonstrates the feasibility and practical value of the proposed method in vocational education contexts, offering a scalable model for aligning software engineering education with industry needs.

This study investigates software development teams’ awareness, attitudes, and readiness for organizational change prior to migrating to Software Product Line (SPL) engineering in Small and Medium-sized Enterprises (SMEs). Using semi-structured, in-depth interviews with key stakeholders across multiple roles, we conducted a qualitative study grounded in an SPL implementation framework and applied thematic analysis. Results indicate unanimous recognition of the migration’s strategic benefits, confirming the critical role of early stakeholder engagement in mitigating transition risks. Based on empirical findings, we propose a three-dimensional strategy to alleviate resistance to change: sustained cross-functional communication, incremental adoption of existing practices, and inclusive, collaborative implementation. This work addresses an empirical gap by systematically assessing pre-migration organizational cognition within SMEs—contextually distinct from large enterprises—and delivers actionable, context-sensitive change management guidance tailored for resource-constrained software organizations undertaking SPL adoption.

This study addresses the core challenges of low B2B sales conversion rates and difficulty in customer expansion. We propose a three-layer closed-loop business intelligence framework: “causal prediction—constrained optimization—generative service.” Methodologically, we introduce a novel technical pipeline integrating causal machine learning (to identify causal mechanisms underlying lead conversion), contextual bandits (for dynamic sales strategy adaptation), and generative AI (to automatically produce personalized outreach content), all orchestrated within a feedback-driven iterative system. Validated on real-world LinkedIn sales data, our approach achieves a 23.6% improvement in lead conversion rate and an 18.4% increase in average customer expansion revenue over conventional methods, while substantially enhancing operational decision interpretability. The framework demonstrates strong cross-industry generalizability, offering a reusable methodology and engineering paradigm for intelligent B2B sales automation.

Inconsistent definitions of “feature” across software engineering domains—particularly requirements engineering (RE) and software product lines (SPL)—impede communication, trigger rework, and reduce cross-team collaboration efficiency. Method: We conducted an empirical study across 27 mainstream open-source projects, integrating repository mining, branch behavior analysis, qualitative coding, and pattern induction to derive a data-driven, cross-disciplinary definition of feature. Contribution/Results: This work introduces the first empirically grounded, unified feature definition framework bridging RE and SPL. It identifies recurring collaboration patterns and critical bottlenecks in feature description, implementation, and management, and proposes a roadmap linking academic theory with industrial practice. The findings yield actionable guidelines for project planning, resource allocation, and inter-team coordination, advancing feature conceptual standardization and engineering practice optimization.

本文提出Phorecaster365架构，通过结合企业资源规划数据和机器学习模型，为药品销售预测提供支持，并强调了人工监督的重要性。

This study addresses the unclear practical impact of generative AI in requirements engineering (RE) within current industrial practice, particularly regarding tool integration, team collaboration, and organizational adaptability. Drawing on a company-wide use case survey conducted in 2024 and two rounds of interviews with eight product owners during 2025–2026, the research systematically analyzes fifteen RE use cases across four categories, leveraging an in-house chatbot and seven commercial generative AI tools. Findings reveal that AI adoption has moved beyond individual productivity gains to influence complex scenarios such as cross-tool integration, customer governance responses, and role boundary reconfiguration. The degree of tool integration critically determines performance benefits, while single-user interaction modes may undermine collaborative dynamics. The study proposes a practitioner-oriented set of evaluation questions to guide effective industrial deployment of AI in RE.

This study addresses the challenge faced by production system engineers in automatically verifying production line layouts due to limited knowledge of PDDL and planning theory. To bridge this gap, the authors propose a novel approach based on an Asset Administration Shell (AAS) capability model that natively generates complete PDDL planning problems directly from domain-level descriptions, eliminating the need for PDDL-specific submodels. The method integrates four Industry 4.0 standards—VDI 3682, IEC 61360-1, IDTA 02011, and IDTA 02016—to construct the AAS and employs an extraction algorithm to automatically translate multi-AAS architectures into PDDL domains. In a laboratory case study, the approach enabled engineers to systematically compare four layout variants by modifying only the AAS model, significantly lowering the barrier to adopting automated planning in industrial settings.

In organizations lacking a formal architect role, software architecture decisions are frequently made by practitioners without such titles. This study employs a mixed-methods approach, combining a survey of 54 practitioners with in-depth interviews of 7 participants, to investigate who actually makes architectural decisions and under what contextual conditions. The findings reveal that informal architects are extensively involved in critical architectural choices, while formally designated architect roles are primarily deemed necessary in large enterprises or complex teams. These results challenge the prevailing research paradigm that centers on formal architects, instead illuminating how architectural responsibilities are distributed across real-world development environments and highlighting their strong dependence on organizational context.

Existing CAD generation models struggle to emulate engineers’ iterative design processes and lack the capability to validate physical and structural compliance. This work proposes an industry-native CAD generation framework that produces complete multi-part STEP files from engineering text and, for the first time, integrates finite element analysis (FEA) into the generative loop to verify structural plausibility. The approach leverages structured blueprint descriptions and 21-view image renderings as dual supervisory signals to guide large language model agents—such as GPT-5.5 and Claude Code—toward self-improving generation. Evaluated on the S2O and Fusion360 datasets, the method significantly enhances geometric reconstruction quality and engineering compliance, improving Box-IoU from 0.444 to 0.592 and from 0.397 to 0.505, respectively.