Ania Cravero

Large language model agents are increasingly used to analyze data, generate code, and execute operations in data-engineering environments. However, successful technical execution does not guarantee that an operation is authorized or that its results are correct and verifiable. This creates governance challenges when specialized agents modify persistent data. This study presents a Governed Human–AI–Agent Workflow that coordinates a Data Expert Agent and a Data Engineering Agent under human supervision. The objective is to establish explicit responsibilities for planning, authorization, execution, verification, and provenance. The workflow was developed using an adapted Design Science Research Methodology and demonstrated through a controlled Data Lakehouse experiment using synthetic academic data. Documented execution failures were analyzed to identify governance requirements and develop operational controls. These failures included repeated execution, inconsistent evaluation units, premature remediation, and unintended effects across data-quality rules. The resulting artifact comprises nine stages and four prompt-based contracts: Planning, Execution, Verification, and Refinement. An initial intervention increased the data-quality detection F1 score from 0.8084 to 0.9077 but introduced 49 false negatives. A subsequent authorized refinement increased F1 to 0.9988, with one remaining false negative. These results illustrate the importance of verifying the effects of technical interventions before accepting a new persistent data state. The contribution is a process-oriented governance artifact that integrates human authorization, specialized agent responsibilities, bounded execution, read-only verification, and persistent evidence. The study provides a formative demonstration within one controlled academic laboratory scenario. Independent evaluation across other datasets, platforms, and institutional environments remains necessary.

Agentic Software Engineering (ASE) systems depend on decisions about context, reasoning, coordination, tool use, execution, and verification. This systematic mapping study maps problems, design decisions, operationalizations, and evidence in peer-reviewed studies of LLM- or foundation-model-based agentic systems in Software Engineering from 1 January 2024 to 13 July 2026. ACM Digital Library, IEEE Xplore, Scopus, and Springer Nature Link were searched on 13 July 2026 using the same Boolean string. From 1069 records, 94 studies were included. A frozen, single-paper-isolated extraction protocol identified 1018 design decisions, followed by deterministic multi-label synthesis; no formal study-level risk-of-bias assessment was performed. Of these decisions, 633 were linked to 656 problems or objectives, and 913 (89.7%) had empirical or comparative evaluation records. Only 19.8% of evaluation records attributed evidence to an individual decision, whereas 52.3% assessed complete systems; multi-agent configurations represented 68.7% of problem-linked decisions. Heterogeneous metrics and system-level evaluation limit inference about effectiveness. The findings expose a gap between evaluation coverage and decision-level attribution and motivate ablations and human-centered evaluations. The review received no external funding and was not registered.

Large language models and AI agents are increasingly changing the way technical work is performed in data engineering. This experience report documents a practical workflow that combined a human researcher, a specialized GPT used as a planning and reasoning agent, and an AI agent integrated into a Lakehouse platform used as an execution agent. The case emerged during a controlled data-governance and data-quality experiment in which the team progressively moved from direct code execution toward a checkpoint-based Human–AI–Agent workflow. The process evolved into a recurring loop of contextualization, planning, instruction, execution, observation, verification, decision, and recontextualization. During this experience, several failure modes were observed, including repeated execution, confusion between physical and logical grain, incorrect rule mappings, referential-integrity violations, and cross-rule side effects introduced by an intervention that improved one metric while degrading recall. These failures motivated practices such as frozen baselines, protected objects, negative instructions, human authorization boundaries, persistent evidence, independent verification, and explicit provenance. The main lesson is that the increasing capability of AI agents does not eliminate the need for technical expertise. Instead, it shifts part of the professional role from direct execution toward orchestration, validation, and accountable decision-making.

Agentic Software Engineering systems depend not only on model choice but also on decisions about context, reasoning, coordination, tool use, execution, and verification. The literature provides a fragmented account of how motivating problems connect to design decisions, operationalizations, and evidence. We address this gap through a systematic mapping study of 94 peer-reviewed primary studies selected from 1,069 records. Evidence-grounded extraction under a frozen, single-paper-isolated protocol identified 1,018 design decisions, of which 633 were explicitly linked to 656 problem or objective mentions. One author audited a random sample of 10 studies comprising 106 decisions and 637 evidence records and found no candidate- or evidence-level discrepancies. Deterministic multi-label codebooks were constructed from the extracted vocabulary and applied using complete token-boundary matching, preserving traceability to source evidence. Although 913 decisions (89.7%) carry an empirical or comparative evaluation record, only 19.8% of evaluation records attribute evidence to an individual decision, whereas 52.3% assess complete systems. Multi-agent configurations account for 68.7% of problem-linked decisions, while hybrid configurations and early life-cycle activities remain sparsely represented. These findings expose a substantial gap between evaluation coverage and decision-level attribution and motivate more targeted ablations and human-centered evaluations.

Large language models and AI agents are extending data-engineering automation beyond isolated artifact generation toward end-to-end processes in which agents interpret requirements, select data, generate transformations, invoke tools, validate results, and communicate analytical outputs. This shift introduces risks that conventional notions of data quality and execution success do not fully capture. A dataset may satisfy established quality standards, and a generated query may execute without technical errors, while the agent still selects an incorrect metric, combines incompatible analytical grains, accesses unauthorized data, or draws conclusions that are insufficiently supported by evidence. This paper develops a conceptual framework for Agentic Data Engineering centered on Quality of Agentic Data Use, defined as the extent to which an agent uses and communicates data in accordance with task, semantic, quality, security, governance, and provenance requirements. An evidence-informed analysis of Data Contracts, Semantic Layers, Data Quality, Guardrails, AI Governance, and Data Provenance shows that these foundations provide essential but fragmented capabilities. The proposed framework integrates and extends them through four core artifacts: Agentic Data Contracts, Agentic Expectations, Agentic Data Provenance, and Agentic Data Governance. It also introduces an execution lifecycle, a reference architecture, a failure taxonomy, and a multidimensional evaluation framework. A governed sales-analysis scenario illustrates how the proposed artifacts interact throughout an agent-mediated data process. In addition, a controlled Databricks prototype and a complementary benchmark comprising 10 cases and 40 executions demonstrate the framework’s technical feasibility and support the independent computation of enforcement indicators. The benchmark highlights the value of separating generation from validation while also showing that the current validation and automated-repair mechanisms require further calibration. These preliminary findings do not establish generalized improvements in safety, correctness, or reliability. Rather, they provide an operational foundation for broader empirical evaluation of trustworthy agent-mediated data-engineering processes.

The integration of large language models and autonomous agents into data engineering is shifting data systems from human-specified pipelines to processes in which agents interpret requirements, select sources, generate transformations, invoke tools, and communicate analytical results. This shift introduces a governance gap: a dataset may satisfy established quality requirements and a generated query may execute successfully while the agent applies an incorrect metric, combines incompatible grains, uses unauthorized sources, or produces conclusions without evidential support. We develop a conceptual framework for Agentic Data Engineering, the engineering domain concerned with specifying, validating, governing, and auditing data processes in which AI agents operate under delegated authority. An evidence-informed analysis of six foundations—Data Contracts, Semantic Layers, Data Quality, Guardrails, AI Governance, and Data Provenance—shows that existing controls are essential but fragmented. The framework integrates them through four artifacts: Agentic Data Contracts, Agentic Expectations, Agentic Data Provenance, and Agentic Data Governance, and defines Quality of Agentic Data Use as the central evaluative construct. A lifecycle model, a reference architecture, a failure taxonomy, a multidimensional evaluation scheme, and an illustrative governed sales-analysis scenario operationalize the framework and provide a foundation for the empirical validation of trustworthy agentic data systems.

The growing strategic importance of data has increased organizational interest in Data Governance (DG) as a mechanism to ensure data quality, security, compliance, and value generation. Despite the widespread adoption of DG initiatives, organizations continue to face substantial implementation difficulties that limit the effective use of data for decision-making and innovation. This study presents a Systematic Literature Review conducted in accordance with the PRISMA framework to identify and synthesize the main challenges associated with DG implementation. A total of 38 empirical studies published between 2017 and 2025 were analyzed. The findings reveal five recurrent dimensions of implementation challenges: organizational, cultural, technical, quality, and regulatory. Frequently reported issues include insufficient resources, shortage of specialized personnel, resistance to organizational change, data silos, interoperability limitations, poor data quality, and increasing regulatory complexity. The review also shows that formal governance frameworks provide implementation guidance but do not fully eliminate these barriers. Based on the synthesized evidence, the study also proposes an interpretive conceptual model to illustrate how these barriers may emerge and reinforce one another during DG implementation. The study concludes that successful DG implementation depends on aligning organizational capabilities, technological infrastructure, and regulatory practices within an integrated governance strategy that supports data-driven value creation.

This systematic review analyzes the evolution of Machine Learning and Big Data applications in agriculture from 2021 to 2025, with particular emphasis on how recent technological advances facilitate the transition from precision agriculture to Intelligent Agricultural Ecosystems. A comprehensive literature search was conducted across Scopus, Web of Science, IEEE Xplore, the ACM Digital Library, SpringerLink, and MDPI, following the PRISMA 2020 guidelines. After duplicate removal and a two-stage screening process (title/abstract screening followed by full-text assessment), eligible peer-reviewed studies were systematically extracted using a structured coding matrix encompassing six analytical domains: crops, soil, weather and water, land use, animal systems, and farmer decision-making. The findings reveal a substantial increase in ML-driven agricultural analytics. Although Random Forest and Convolutional Neural Networks remain widely adopted, recent studies demonstrate a marked shift toward advanced Deep Learning architectures, integrated cloud–edge–device infrastructures, Federated Learning frameworks for privacy-preserving collaboration, Explainable AI techniques to enhance transparency, and governance-oriented mechanisms to ensure interoperability. Notwithstanding these advances, several persistent challenges remain, including limited generalizability across diverse agroclimatic contexts, the high costs associated with high-quality data annotation, the integration of heterogeneous and multimodal datasets, and infrastructural constraints related to connectivity. These developments are synthesized within the IAE conceptual framework, underscoring governance- and lifecycle-aware orchestration MLOps as a critical differentiator that transcends purely technology-centric approaches.

Meta4CBC is proposed as an innovative metamodel for competency-based curriculum design in higher education, addressing the critical need to adapt curricula to contemporary challenges driven by globalization, technological advancement, and climate change. Inspired by the Business Motivation Model and Semantics of Business Vocabulary and Business Rules metamodels, Meta4CBC seeks to overcome the limitations of current CBCD models by providing a coherent structure that integrates curriculum components at supra, macro, meso, micro, and nano levels and establishes a common language that facilitates interdisciplinary collaboration. The proposal promotes curriculum alignment and coherence while supporting the development of software tools to assist in curriculum design, automate repetitive tasks, and provide data analysis for informed decision-making. The validation of Meta4CBC was conducted through a multi-step methodology involving proof of concept, expert judgment, and practical application. After defining the components through a systematic literature review, the metamodel was tested in the Computer Engineering program at the University of La Frontera. This testing verified its adaptability and alignment across curriculum levels. Experts from various Chilean universities reviewed the model, providing feedback for refinement.

Context: Quantum software development is a complex and intricate process that diverges significantly from traditional software development. Quantum computing and quantum software are deeply entangled with quantum mechanics, which introduces a different level of abstraction and a deep dependence on quantum physical properties. The classical requirements engineering methods must be adapted to encompass the essential quantum features in this new paradigm. Aim: This study aims to systematically identify and analyze challenges, opportunities, developments, and new lines of research in requirements engineering for quantum computing. Method: We conducted a systematic literature review, including three research questions. This study included 105 papers published from 2017 to 2024. Results: The main results include the identification of problems associated with defining specific requirements for quantum software and hybrid system requirements. In addition, we identified challenges related to the absence of standards for quantum requirements engineering. Finally, we can see the advances in developing programming languages and simulation tools for developing software in hybrid systems. Conclusions: This study presents the challenges and opportunities in quantum computing requirements engineering, emphasizing the need for new methodologies and tools. It proposes a roadmap for future research to develop a standardized framework, contributing to theoretical foundations and practical applications.