Serkan Kumyol
AI Consultant & Ph.D. Candidate
For eight years I have been building AI systems at HKUST that span clinical speech processing, multi-agent simulation, generative narrative tools, and enterprise document pipelines — always with a formal evaluation layer, because the difference between a demo and a deployed system is measurable. That work has been published at ACM Creativity & Cognition, Interspeech, and IEEE, tested with real users, and validated on production document workflows. I am finishing my Ph.D. in Computer Science in August 2026.
The thread running through all of it goes back further. Before transformers existed, my master's thesis was a mathematical attempt to learn the hidden structure of Turkish morphology from data alone — a language where a single word compresses what English spreads across an entire clause. It taught me to think about language as compressed structure rather than text, and to ask what a system actually learns rather than what it appears to do. That question has not changed; only the systems have.
Based in Hong Kong since 2017, looking for an applied role where the research question has a real deployment target.
Experience
AI Analyst at dRoW Limited
Evaluated and fine-tuned OCR/LLM options on enterprise document datasets. Built local inference pipelines with vLLM and Ollama. Designed model-quality framework and implemented Python/PyTest regression checks.
Research Assistant at HKUST
Designed and delivered applied AI systems from concept to working prototype. Built NarrativeHive (multi-agent LLM + RAG), Orchid, SeaSense, and HK-GenSpeech. Mentored 30+ students per semester in NLP/ML labs.
Software Engineer at Arskom Group
Built mission-critical billing software (Python, Django, SOAP/RPC), improving billing accuracy by ~40%. Reduced system error rates by ~75% through QA automation.
Selected Work
See all projectsSelected projects across speech, generative AI, interactive systems, and applied machine learning.
Generative AI + Wellbeing • 2024
SeaSense
Turns written emotions into AI-grown 3D flowers, transforming a public screen into a shared emotional garden.
Healthcare AI + Speech • 2023
AlzDetect (HK-GenSpeech)
Reimagines dementia screening as a richer conversation, using culturally local image prompts and speech modeling to estimate cognitive scores.
Interactive Narrative + LLMs • 2022
Orchid Narrative System
A graph-based authoring environment that lets LLM-driven stories grow freely while still climbing a deliberate narrative trellis.
Multi-Agent Systems + Storytelling • 2021
Narrative Hive
A multiplayer storytelling world where humans and AI characters improvise together, but with enough memory and governance to keep the society coherent.
Publications
These papers are the written trail of the same journey: from morphology and machine translation to creative systems, multimodal interaction, and applied healthcare AI.
Morphological Segmentation and Bayesian Models for Turkish Language Processing
Kumyol, Serkan
Modeling morpheme triplets with a three-level hierarchical Dirichlet process
Kumyol, Serkan and Can, Burcu
Allomorphs and binary transitions reduce sparsity in turkish semi-supervised morphological processing
Can, Burcu and Kumyol, Serkan and Bozsahin, Cem
HK-GenSpeech: A Generative AI Scene Creation Framework for Speech Based Cognitive Assessment
Yong, Vi Jun Sean and Kumyol, Serkan and Low, Pau Le Lisa and Leung, Suk Wai Winnie and Braud, Tristan
Myokey: Surface electromyography and inertial motion sensing-based text entry in ar
Kwon, Young D and Shatilov, Kirill A and Lee, Lik-Hang and Kumyol, Serkan and Lam, Kit-Yung and Yau, Yui-Pan and Hui, Pan
Efficient Bilingual Generalization from Neural Transduction Grammar Induction
Yan, Yuchen and Wu, Dekai and Kumyol, Serkan
Orchid: A Creative Approach for Authoring LLM-Driven Interactive Narratives
Wu, Zhen and Kumyol, Serkan and Wong, Shing Yin and Hu, Xiaozhu and Tong, Xin and Braud, Tristan
Technical Skills
AI & Machine Learning
- PyTorch
- TensorFlow
- Hugging Face
- scikit-learn
- LangChain
- vLLM
- Ollama
- RAG
- Agentic Workflows
Programming & Data
- Python
- SQL
- C++
- JavaScript/TypeScript
- Bash
- Pandas
Cloud & Systems
- AWS
- GCP
- Azure
- Docker
- GitHub Actions
- Jenkins
- Linux/Unix
Education
My academic background combines computer science, cognitive science, and education. That combination shaped both how I build AI systems and how I explain them.
Ph.D. in Computer Science
Hong Kong University of Science and Technology (HKUST)
Research focused on applied AI systems, including speech processing, multi-agent interaction, generative tools, and user-facing evaluation.
M.Sc. in Cognitive Science
Middle East Technical University (METU)
Studied language, learning, and perception, with thesis work on unsupervised morphological learning in Turkish.
B.Sc. in Computer Systems & Education
Gazi University
Built a foundation in computing while also training in how to teach technical material clearly and systematically.