Hi there 🌟, my name is Natthanan Bhukan, or you can call me "Tae". Currently, I am working as a machine learning engineer at CJ Express Group to develop a machine learning system in the retail sector. I graduated from King Mongkut's University of Technology Thonburi.
My research interests focus on developing efficient system support to optimize AI applications within the system and algorithm.
Journal of Food Engineering (Being review)
Machine Learning Engineer
[Part time] Teacher assistant in image processing and computer vision (CPE463)
AI Engineer
Teacher assitance for Data Science Essential class
About monitoring and observability in LLM applications, focusing on Langfuse, a tool that helps track system behavior, prevent issues, and enhance future performance for LLM projects.
Explore the unique challenges of serving LLMs, understnding their techniques, and demonstrate solutions using vLLM.
Recap my presentation in DevFest Cloud Bangkok 2023 which show how to integrate of Ray with VertexAI, known as Ray on VertexAI, and demonstrate its application in LLMOps within the VertexAI ecosystem.
Introduce Yatai, a machine learning platform for scalable model deployment. In this article, discuss its features, its role in the MLOps lifecycle, and provide a demo showcasing its capabilities.
This talk will look at KubeRay, an open-source Kubernetes operator designed to simplify the deployment and management of Ray clusters for distributed computing.
In the rapidly evolving field of artificial intelligence, the deployment of large language models (LLMs) presents both immense opportunities and complex challenges. To ensure responsible and effective innovation, it is crucial to have a comprehensive understanding of how to improve LLM applications to solve business use cases.
Large language models (LLM) are rapidly growing, and numerous applications have emerged. Many of these applications deal with a quick change in data and business requirements, which may cause a problem in the system.
This project provides a high-performance implementation of YOLOv11 object detection using TensorRT for inference acceleration. The pipeline processes images and videos in batches, leveraging CUDA for preprocessing and inference. Non-Maximum Suppression (NMS) and postprocessing are performed on the CPU to optimize results.
A pioneering open-source project providing an efficient, predictive personal protective equipment (PPE) detection system. This revolutionary technology leverages high-performance object detection models by using TensorRT pair with orchestrated services and asynchronous tasks handling.
a web application base that help you to classify the grabage by using AI. We want you to handle the grabage easier, so we come up with this solution to help you. This project is part of one night miracle team
GitHub
LinkedIn
CV