cv

PhD Student

University of Rochester

Computer Science

• End-to-end Deep Learning

Master of Engineering

Xi'an Jiaotong University

Computer Science and Technology

• Big Data and Deep Learning
• Computer Vision

Bachelor of Engineering

Institut Teknologi Sepuluh Nopember

Engineering Physics

• Neural Networks

Senior Data Scientist

GoTo Group

Working on eKYC and computer vision projects.

Machine Learning (ML) Engineer

Fintelite

Developing multi-modal LLM-based key information extraction systems.

• Integrated multi-modal LLM for document key information extraction.
• Optimized prompt templates for efficient extraction.
• Enhanced dataset annotation platform with new tools.
• Introduced dataset versioning using Hugging Face datasets.
• Developed document-agnostic key information extraction system.

Artificial Intelligence (AI) Engineer

Nodeflux

Developing and optimizing AI systems for real-time computer vision analytics.

• Developed end-to-end AI systems for real-time analytics.
• Optimized deep learning models for face and vehicle recognition.
• Improved LPR pipeline efficiency and accuracy.
• Explored novel object detection architectures.
• Implemented active learning techniques for object detection.

Data Curriculum Program Lead

Data Science Indonesia

Developed a comprehensive data-focused curriculum for data scientists, engineers, and analysts.

• Led a team of 11 experts to strategize and develop the curriculum.
• Created a GitHub repository with over 160 stars and a GitBook learning roadmap.

Evolving Parameterized Prompt Memory for Continual Learning

AAAI Conference on Artificial Intelligence

Research on evolving parameterized prompt memory techniques to improve continual learning.

Feature Space Topology-Preserving Transfer for Anomaly Detection

Pattern Recognition Letters

Study on preserving feature space topology to improve anomaly detection performance.

Contextual Prompting for Continual Learning

Master Thesis

Master thesis exploring contextual prompting strategies for continual learning.

Online Continual Learning via Multiple Deep Metric Learning and Uncertainty-guided Episodic Memory Replay

ArXiv

Preprint discussing multiple deep metric learning techniques for online continual learning.

Optimization of Neural Network Based on Hybrid Method of Genetic Algorithm and Particle Swarm Optimization for Maritime Weather Forecasting in Buoyweather Station Type II

AIP Conference Proceedings

Optimization study combining genetic algorithm and particle swarm optimization for weather forecasting.

Design of Mobile Predictor for Maritime Weather Forecasting Using Hybrid Method of Type-2 Fuzzy Logic and Artificial Neural Network with Differential Evolution Algorithm Optimization

Bachelor Thesis

Thesis on designing a mobile predictor for maritime weather forecasting using hybrid AI methods.

3rd place in Continual Learning Challenge at ICCV 2021

Chinese Government Scholarship (CSC) for Master Study