"Exploring how machines perceive and reason — uniting vision, language, and intelligence into meaningful understanding."
I am an AI/ML Engineer and master's student at Institut Teknologi Bandung (ITB), Indonesia, specializing in deep learning for multimodal perception and language understanding. My work spans computer vision, natural language processing, and knowledge-augmented AI systems, with a focus on vision-language modeling, crossmodal retrieval, and representation learning.
My current thesis at ITB's AI program focuses on Knowledge Distillation for 3D Object Detection, bridging high-performance multimodal models (camera + LiDAR) with real-time edge deployment — a core challenge in autonomous driving and robotics.
Riset AI
Autonoma
Coding Camp powered by DBS Foundation
Dicoding DBS
Designed a pipeline comparing Swin Transformers vs CNNs for brain MRI classification. Swin-Base achieved 96.2% accuracy, outperforming baselines.
Generated synchronized LiDAR + camera samples in CARLA. Trained BEVFusion and built a web dashboard for 3D bounding box visualization.
Integrated PubMed embeddings, vector search, and Neo4j for semantic QA over biomedical entities and relations for explainable discovery.
Trained YOLOv5 on custom-labeled images for sharp-turn vehicle safety detection. Deployed on Raspberry Pi for real-time edge inference.
Implemented YOLOv4 on AWS for monitoring broiler chickens to reduce disease spread. Deployed end-to-end with a real-time web dashboard.
Collected sensor data through API, stored in MongoDB, and visualized via React.js front-end for future research analytics.
Programming Languages
ML / DL Frameworks
AI Domains & Specializations
MLOps & Tools