Dewi Yanti Liliana
IEEE Senior Member
Department of Computer and Informatics Engineering, Politeknik Negeri Jakarta (PNJ)
Artificial Intelligence for Industry and Applications (AIRA) Center of Technology, PNJ
Biography
Dr. Dewi Yanti Liliana is an Associate Professor in the Department of Computer and Informatics Engineering at Politeknik Negeri Jakarta (PNJ), Indonesia. Her research focuses on artificial intelligence, computer vision, and responsible AI, with an emphasis on translating research into practical solutions for industry and society. She contributes to applied AI innovation through the AIRA Center of Technology (Artificial Intelligence for Industry and Applications) at PNJ. She is an IEEE Senior Member and has served on the IEEE Region 10 Women in Engineering (WIE) Committee since 2022. Her international research and academic engagement include research collaboration supported by the British Council (2024) and participation in the Australia Awards Scholarship program at RMIT University (2025). She also serves as the 2026 Chair of IEEE Indonesia Section Women in Engineering, supporting women’s leadership and participation in science, technology, and engineering.
Abstract
AI is changing the way we work with information and solve problems. While generative AI has made it easier to create content and obtain answers, the next step is to develop AI systems that can plan, use tools, and carry out tasks with greater autonomy. As these systems take on more complex responsibilities, questions of trust, reliability, and ethics become increasingly important. An effective Agentic AI system must do more than complete a task; it must also produce verifiable results, operate within clear boundaries, and remain accountable for its actions. Achieving this requires careful system design, combining large language models with specialized agents, external tools, structured data, and reliable computational processes. One example is metadata-centric multi-agent AI analytics, where coordinated agents use structured metadata and deterministic computation to analyze large datasets more efficiently. Beyond this technical approach, the broader challenge is to ensure that autonomous AI systems are transparent, appropriately supervised, and designed to minimize errors, bias, and unintended consequences. Balancing capability with responsibility will be essential for bringing Agentic AI into research, industry, and other real-world settings where trust matters as much as performance.