Building trustworthy multimodal AI for human behavior, complex systems, and research-based decision support — with a focus on the Indonesian context and Global South perspectives in AI.
"I develop trustworthy multimodal AI systems validated across human behavioral contexts and complex real-world systems — advancing interpretable, fair, and accountable AI for the Global South, with a particular focus on the Indonesian language, cultural, and social context."
Building systems that capture behavioral signals from visual, audio, and textual modalities through graph-based and spatio-temporal modeling.
Going beyond accuracy — ensuring models are explainable, fair, robust, and accountable through graph-theoretic interpretability mechanisms.
Human-centered AI and behavioral intelligence — deception cue analysis, political discourse, and cognitive load detection for forensic decision support.
Generalizing trustworthy multimodal AI beyond human behavior — maritime safety, graph optimization, and operational decision support.
Peer-reviewed publications in multimodal AI, deception detection, and behavioral computing.
Research-oriented consulting services for academics, researchers, and institutions.
Literature review, systematic review mapping, research proposal refinement, and manuscript structure review.
Data cleaning, exploratory analysis, visualization, and Python-based statistical workflows for academic research.
Audio preprocessing, acoustic feature extraction, speech-to-text pipeline, and voice-based research workflows.
Interpretability review, bias & fairness assessment, culturally-aware AI evaluation, and responsible AI documentation.
Conference slides, research posters, technical reports, academic CV, and grant proposal support.
Model evaluation, classification experiments, feature engineering, and research-oriented ML workflows.
Insights on AI research, academic life, and the journey of building trustworthy technology.
A new space to share research, reflections, and the journey of building trustworthy AI in the Indonesian context.
Ruang baru untuk berbagi riset, refleksi, dan perjalanan membangun AI yang dapat dipercaya dalam konteks Indonesia.
Why high-accuracy AI systems can still fail — and why trustworthiness must be the new standard.
I am actively seeking interdisciplinary collaborators and open to selected consulting projects. Let's build something meaningful together.