I'm a Data Scientist, AI Engineer, ML Engineer, and Geospatial Data Scientist based in Southeast Asia (Singapore/Jakarta). My passion lies in extracting insights from complex datasets, building intelligent systems, and solving real-world problems using machine learning and artificial intelligence, including a focus on geospatial data.
- 🔭 I’m currently working on building Generative AI/Agentic AI solution in various industries in Asia Pacific
- 🌱 I’m currently learning the intersection of Geospatial Data Science and Generative AI
- 👯 I’m looking to collaborate on open-source AI/ML projects, geospatial data analysis, or impactful data science initiatives.
- 💬 Ask me about machine learning, Generative AI, deep learning, geospatial data processing, Python for data science, and data visualization.
- 📫 How to reach me: [[email protected]]
Here are some of the technologies and tools I work with:
Programming Languages:
Data Science & ML Frameworks:
Geospatial Libraries & Tools:
Databases & Big Data:
Cloud & MLOps:
Visualization & Tools:
Here's a curated list of some of my projects, categorized by their focus. Feel free to explore the repositories for more details!
| Project Name | Description | Technologies Used | Link |
|---|---|---|---|
| Streaming Multi Task Vision Agent | A streaming vision agent that can handle multi-tasks. This agent connect to RAG through GCP RAG Engine, vector datastore, or ChromaDB | Python, Google ADK, Google Cloud Service, ChromaDB | Repo Link |
| Multimodal RAG Chatbot Agent | A Multimodal Chatbot Agent which support Text and Audio as input by empowering Speech-to-Text (STT) and Text-to-Speech (TTS) in Azure Opean AI Service. The chatbot also is built using RAG pipeline for grounding the contextual information. | Python, Azure Open AI, Langchain, Huggingface | Repo Link |
| Azure Search - RAG Chatbot | A RAG chatbot using Azure Search as RAG. | Python, Azure Open AI, Azure Search Langchain, Huggingface | Repo Link |
| Azure - MCP | An Model Context Protocol (MCP) in Azure. | Python, Azure Open AI | Repo Link |
| Multilingual RAG-Chatbot Agent | Create chatbot agent using Azure Open AI & RAG pipeline. | Python, Azure Open AI, Langchain, Huggingface | Repo Link |
| Google Gemini CV Analyzer | Building multimodal CV analyzer using Google Gemini (Automated Resume Tracker). | Python, Google Gemini, Cloud Run | Repo Link |
| Realtime Voice Agent | Real Time Voice Agent Using Gemini (LLM), Google Text-to-Speech (TTS), and Google Speech-to-Text (STT). | Python, Google Gemini, Google TTS, Google STT, Cloud Run | Repo Link |
| Agent for Booking & Scheduling | Using Gemini model to create agent for scheduling and booking. | Python, Google Gemini | [Repo Link] |
| Gemini for OCR | Leveraging Google Gemini as foundational model for OCR application | Gemini | Repo Link |
| Agent & LLM Evaluation | Evaluate the performance of LLM/Agent using Opik Comet (hallucination metrics, usefulness, relevance). | Opik Comet | Repo Link |
| Project Name | Description | Technologies Used | Link |
|---|---|---|---|
| End-to-End ML Modeling & Deployment | This is an end-to-end machine learning project, started from model development (supervised model development), pipeline for model development and GridSearchCV to find the best hyperparameter, MLOps using MlFlow, create app for model deployment using Flask, and app deployment using Docker | Python, Flask, Docker, MLOps, MLFlow, Scikitlearn, XGBoost | Repo Link |
| Apache Spark ML | Machine Learning Model Development using Big Data Framework Apache Spark ML | Python, Apache Spark, SparkML | Repo Link |
| ML Engineering | Machine Learning Engineer (MLE) that covers end-to-end ML project (model development, CI/CD, model deployment. Focus on project deployment using Docker and Kubernetes in Google Cloud GKE. | Python, Docker, Github Actions (CI/CD), Kubernetes, Scikit-learn, Kubeflow, Flask | Repo Link |
| Apache Airflow ML Pipeline | Building ML Pipeline using Apache Airflow. | Python, TensorFlow, Docker, Apache Airflow | Repo Link |
| Project Name | Description | Technologies Used | Link |
|---|---|---|---|
| Image Classification using Transfer Learning | Transfer learning techniques using popular pre-trained models: Inception, ResNet, MobileNet, and EfficientNet. | Python, Tensorflow, Transfer Learning | Repo Link |
| Multiclass-Image Segmentation Unet-Resnet34-VGG16 | This project aims to create a multiclass image segmentation from aerial imagery using Unet transfer learning Resnet34 and VGG16. The image contains information about land cover (built-up area, water bodies, vegetation, and miscellaneous) | Python, Tensorflow, Transfer LearningL | Repo Link |
| Unet_TransferLearning_Resnet34 | The project aims to enhance the accuracy of palm oil plantation segmentation (aerial imagery) by integrating the U-Net architecture with transfer learning from ResNet-34. | Python, Tensorflow | Repo Link |
| Deep Unet | Building a Deep Unet Architecture to Make Image Semantic Segmentation for Building Area from Satellite Imagery. | Python, TensorFlow | Repo Link |
| Project Name | Description | Technologies Used | Link |
|---|---|---|---|
| Climate Automation Modeling | Project software engineering to create automation workflow to proceed climate data into geospatial data. This project covers end-to-end from ETL, data processing using Python/ArcPy/Geoprocessing, dashboard visualization, and report (maps, infographics, short report) generating.. | Python, Arcpy, ArcGIS, GIS | Repo Link |
| Sentinel 2 Image Classification | Machine Learning Application for Image Classification (Sentinel 2). | Python, Tensorflow, XGBoost, Scikit-learn, numpy | Repo Link |
I'm passionate about contributing to the broader scientific and technical community. Here are some of my research papers and publications:
| Title | Publication/Conference | Year | Link |
|---|---|---|---|
| Machine Learning and Bayesian Method For Monitoring and Forecasting Jayawijaya’s Tropical Glacier Change (Papers Track) | International Conference on Learning Representations (ICLR) | 2025 | Paper Link |
| Comparison of machine learning land use-land cover supervised classifiers performance on satellite imagery sentinel 2 using lazy predict library | Indonesian Journal of Data and Science | 2023 | Paper Link |
| Spatial multi-criteria analysis for urban sustainable built up area based on urban heat island in Serang City | IOP Conference Series: Earth and Environmental Science | 2019 | Paper Link |
| Onshore oil and gas reservoir detection through mapping of hydrocarbon microseepage using remote sensing | IOP Conference Series: Earth and Environmental Science | 2019 | Paper Link |
| Hydrocarbon Microseepage Potential Area Exploration Using Sentinel 2 Imagery | E3S Web Conf. | 2018 | Paper Link |
| Routing the highway development by using SuperMap Least Cost Path Analysis (LCPA) and Multi-Criteria Decision Analysis (MCDA) and its assessment toward spatial planning | IOP Conference Series: Earth and Environmental Science | 2019 | Paper Link |
Made with ❤️ by Iqbal