Shouvik
Ahmed Antu

SHOUVIK

I build conversational AI that talks with students, in text and in voice, and measure whether it actually improves how they reason. Now an Erasmus Mundus MSc student in AI at Universitat Pompeu Fabra, Barcelona.

NOW
Erasmus Mundus MSc in AI, UPF Barcelona and Sapienza Rome, 2026–28
LAB
Research assistant, TIDE Lab, Universitat Pompeu Fabra
NEXT
Presenting at SIGCSE TS, Sacramento, Feb 2027

Publications

  1. 2027
    Let Them Think Out Loud: Designing Voice AI for Reasoning and Reflection in Computing CoursesAntu, S. A., Cordova, L.ACM SIGCSE Technical Symposium 2027 (Accepted)
    SIGCSE TS
  2. 2025
    Engaging with the Wiphala: A Code-Generation Hardened CS1 FinalDeutschbein, C., Ahmed Antu, S.Journal of Computing Sciences in Colleges
    JCSC
  3. 2023
    Using LLM (Large Language Model) to Improve Efficiency in Literature Review for Undergraduate ResearchAntu, S., Chen, H., Richards, C.AI4ED Workshop, CEUR-WS Vol. 3487
    CEUR-WS
  4. 2023
    An Exploration Into the Collatz Conjecture with Changed ParametersAntu, S. A., Shrimali, R., Jones, M.arXiv:2305.01017 (Preprint)
    arXiv

Featured project

Voice AI · AI4Ed Research Lab

Continuity

Students explain course concepts out loud to an AI peer that asks follow-up questions. Professors publish the topics and reference material, and every conversation becomes research data on how students reason.

Design lessons from deploying it in a Willamette computing course were accepted to SIGCSE TS 2027.

Research experience

  1. Oct 2026 – Present
    Research AssistantTIDE Lab, Universitat Pompeu Fabra · Barcelona, Spain
    • Research conversational AI agents for collaborative learning.
    • Build Python research software that turns qualitative coding of educational interaction data into reproducible quantitative pipelines.
    • Apply learning analytics and network-based methods to study collaboration in technology-enhanced learning.
  2. May 2025 – Nov 2026
    Research AssistantAI4ED Lab, Willamette University · Salem, Oregon
    • Studied how non-evaluative, voice-first dialogue improves student reasoning and brings confusion to the surface sooner.
    • Designed a modular prompt architecture with dynamic personas, feedback modes and system-level behavioral controls.
    • Built automated scoring of reasoning quality from voice transcripts: evidence use, trade-off analysis and argument structure.
  3. Jun 2025 – May 2026
    CS Research AssistantVision Lab, Willamette University · Salem, Oregon
    • Worked with hybrid deep learning and classical computer vision models for image alignment, including homography estimation and optical flow.
    • Tested multimodal PyTorch and TensorFlow pipelines that combine RGB and frequency-domain data on digital media.
    • Explored image forensics models that localize manipulation from JPEG and DCT artifacts.
  4. Aug – Dec 2024
    Research AssistantWillamette University Business School · Salem, Oregon
    • Studied stated collaborator preferences in LLMs as a foundation for predicting agentic social decision making.
    • Brought behavioral-science methods to LLM testing.
    • Wrote scripts to vary model parameters systematically.
  5. May – Aug 2023
    SCRP Research Assistant, Generative AISCIS, Willamette University · Salem, Oregon
    • Co-authored research on an AI-driven teaching tool to reduce instructor burnout and increase student engagement.
    • Described the tool's architecture, using LLMs for AI-mediated conversations and automated assessment.
    • Helped design adaptive AI agents and a RAG framework for personalized learning paths.

Talks

  1. Feb 2027
    Let Them Think Out LoudACM SIGCSE Technical Symposium · Sacramento, California
    Upcoming
  2. Oct 2025
    Engaging with the WiphalaCCSC Rocky Mountain · Utah Valley University, Utah
  3. Apr 2024
    AI in Education: Empowering Instructors, Engaging LearnersStudent Scholarship Recognition Day · Willamette University
  4. Oct 2023
    Large Language Models in Higher Education: Applications and UsesTechbytes, SCIS · Willamette University
  5. Sep 2023
    Using LLMs to Improve Literature Review EfficiencyAI4ED Workshop · Tokyo, Japan (virtual)

Teaching

  1. Jan 2019 – Present
    Private TutorIndependent · Remote
    • Tutor high school and undergraduate students in physics, CS1, CS2, algebra, precalculus, calculus and statistics.
    • Use Python examples to make physics and statistics concepts concrete.
  2. Aug 2025 – May 2026
    Section Leader, CS1Willamette University · Salem, Oregon
    • Led weekly lab sections for 8–10 introductory computer science students.
    • Tutored one-on-one, helped debug code and staffed the student help center.
    • Graded projects with detailed feedback, calibrated with faculty in weekly prep meetings.
  3. Aug – Dec 2024
    Colloquium Associate, Thinking MachinesSCIS, Willamette University · Salem, Oregon
    • Led discussions and programming exercises on AI and machine learning concepts.
    • Held office hours on computational thinking and its social implications.
    • Developed course materials and led lab and CHASE sessions with the professor.

Education

  1. Sep 2026 – Aug 2028
    Universitat Pompeu FabraErasmus Mundus Joint Master in Artificial Intelligence (EMAI)Barcelona, Spain · Rome, Italy · Full Erasmus Mundus scholarship · Specialization in Intelligent Decision Making · Mobility at Sapienza University of Rome
  2. Aug 2022 – May 2026
    Willamette UniversityBS in Computer Science and Data ScienceSalem, Oregon · Summa cum laude, GPA 3.9 · School of Computing and Information Sciences
  3. Jan – Jun 2025
    University of BirminghamStudy abroad, Computer ScienceBirmingham, England · GPA 4.0 · Security and Systems, Natural Language Processing, Intelligent Interactive Systems

Awards and honors