My research approach in the field of artificial intelligence is grounded in interdisciplinary thinking, combining rigorous analytical methods with a systems-level perspective. I aim to bridge theory and practice, always aligning my work with real-world challenges and long-term human-centered outcomes.
1. Research Philosophy
I follow a problem-first methodology: I start by clearly defining a high-impact, underexplored question with both theoretical and applied relevance. My process emphasizes clarity, precision, and depth — preferring fewer but more meaningful experiments that lead to concrete insights or frameworks.
This approach is inspired by both scientific inquiry and design thinking, though in research I lean heavily into structured hypothesis testing, reproducibility, and long-term scalability of findings.
2. Tools and Technologies
Depending on the research goal, I work with:
Transformer-based models (e.g., LLMs, BERT, GPT) for language tasks
Clustering, dimensionality reduction, and hybrid methods for exploratory phases
XAI (Explainable AI) tools to interpret and communicate results
Jupyter + Git for open, reproducible code environments
Academic writing tools such as Overleaf, Zotero, and LaTeX for collaborative publishing
3. Long-Term Goals and Mindset
While many projects focus on short-term performance benchmarks, my long-term vision is to build frameworks, tools, and methodologies that can be reused and extended by others. I strive to write in a way that is not only publishable but useful — to other researchers, practitioners, and even policy-makers.
My personal mindset emphasizes:
Intellectual honesty over results chasing
Curiosity-driven exploration
Methodological rigor and reflection
Clear communication of complexity
Go Back
