Researcher

Priyabanta Sandulu

Priyabanta Sandulu is a researcher at the German Research Center for Artificial Intelligence (DFKI) who has mostly used artificial intelligence to model semantic robotic process automation. He studied for a B.Tech in Instrumentation and Control Engineering at Siksha 'O' Anusandhan University in India from 2008 to 2012 and for an MSc in Data Science at Universität Trier from 2021 to 2024, and he worked at Tata Consultancy Services and Aviva Insurance as a software test automation engineer and test data management consultant. In 2025, he co-authored "Actionable Trustworthy AI with a Knowledge-based Debugger" with Andrea Sipka, Sergey Redyuk and Sebastian J. Vollmer, and in October 2026 he co-authored GAIDA, a research prototype for human-in-the-loop document annotation. At DFKI's Department of Data Science and its Applications, headed by Professor Sebastian Vollmer, he is a researcher.

Now at
DFKI

Timeline

  1. 2008–2012

    Engineering degree in India

    From 2008 to 2012, he studies Instrumentation and Control Engineering at Siksha 'O' Anusandhan University in India for a B.Tech.

  2. 2021–2024

    MSc in Data Science

    He studies data science at Universität Trier from 2021 to 2024, for an MSc.

  3. By 2025

    Test automation in India and the UK

    Before joining DFKI, he works at Tata Consultancy Services and Aviva Insurance in India and the UK as a software test automation engineer and test data management consultant.

  4. 2025

    A knowledge-based debugger

    In 2025, he co-authors "Actionable Trustworthy AI with a Knowledge-based Debugger" with Andrea Sipka, Sergey Redyuk and Sebastian J. Vollmer.

  5. October 2026

    The GAIDA annotation prototype

    GAIDA, a research prototype for human-in-the-loop document annotation, is posted to arXiv in October 2026 with Islam Mesabah, Ali Elhosary, Darko Obradovic and Sebastian J. Vollmer as his co-authors. It joins OCR-based extraction, LLM-based semantic pre-annotation and human validation in one editable workflow, and OCR plus LLM pre-annotation with a human reviewer cuts document-annotation time by up to 80%. The work also yields AraReceipt, a dataset of 100 Arabic receipts with 25 key-information classes.

  6. As of October 2026

    Researcher at DFKI

    At the German Research Center for Artificial Intelligence (DFKI), he is a researcher in the Department of Data Science and its Applications, headed by Professor Sebastian Vollmer. His interests are robotic process automation, natural language processing and experience-based learning systems, and he has mostly used artificial intelligence to model semantic RPA processes. He lists AI governance as his research area.

Career

  1. NowDFKIResearcher
  2. Tata SonsSoftware test automation engineer
  3. Aviva InsuranceTest data management consultant

Education

  1. 2024University of TrierMSc
  2. 2012Siksha O Anusandhan UniversityB.Tech

Sources and credits

Photo: We have no published photo of Priyabanta Sandulu yet. Send us a free-licensed photo.

  1. Priyabanta Sandulu — Data Science and its Applications dsa.dfki.de. Checked 22 October 2026.
  2. Actionable Trustworthy AI with a Knowledge-based Debugger — Data Science and its Applications dsa.dfki.de. Checked 22 October 2026.
  3. GAIDA: Generative AI for Human-in-the-Loop Document Annotation arxiv.org. Checked 22 October 2026.
  4. Islam Mesabah · PhD student, RPTU & DFKI islammesabah.github.io. Checked 22 October 2026.
  5. Sandulu Priyabanta - Google Scholar scholar.google.com. Checked 22 October 2026.