I'm a passionate software developer specializing in web development and cloud technologies.
- Programming Languages: Java, JavaScript, Python, Go
- Frameworks:Spring Boot, Angular, React, Django
- Tools: Docker, Kubernetes, AWS
-
Construct Open AI's CLIP model into a docker image, and use it to construct an image search engine, using a cloud-native development method. My model image is public and everyone is welcome to use it. This system mainly demonstrates the basic architecture of an AI-based image search engine. If it is a production environment, please use a vector search engine for optimization, such as Elastic Search, Faiss.
-
A real-time trading system that taught me a lot. Spring Cloud is used to implement the microservice architecture,kafka message queue implements decoupling between services, redis implements transaction snapshots, API gateway implementation, websocket implements server message push, and we can write our own ORM framework.
-
An online library sounds very cool, but what is more interesting is that this project will let us learn more, how to quickly search documents? ElasticSearch is undoubtedly the first choice. The flashback indexing mechanism allows us to quickly find relevant documents through keywords. The only problem is, it's too heavy (local deployment puts my laptop through the paces). Fortunately, we can experience Elastic Cloud for 7 days to explore Elastic Search and Kibana. As well as logstash and various connectors to ensure data synchronization.
-
Spark is a very important framework for data engineering and is divided into two versions: python and scala. The new version of Spark already supports the use of dataframes, so it is not difficult for people who are familiar with Pandas to get started. This project gave me more experience in various deployment methods. To deploy in an AWS environment we can use EKS and EMR, as well as Glue if you want to develop directly with pySpark. It should be noted that EMR cannot configure virtual machines, but directly submits project packages, so the project cannot contain operations that interact with the operating system.
-
A very popular Apache incubator open source project, a few months ago I started trying to contribute to the open source project. I am very happy to choose this project as the first open source project I contributed to. The project's workflow is very standardized and the community members are friendly, which allowed me to learn a lot about code specifications and architectural design. This project is a typical web project with separate front and back ends, developed using React and Golang. Golang, as the current main language for cloud native development, has given me a good experience. If you are already proficient in Java, C, and Python, learning Golang will be very fast. And the backend adopts the MVC architecture (the framework Gin itself is a lightweight framework, so there is no MVC architecture, but the project manually implemented the architecture. There are many such examples. The MVC architecture does a great job in service decoupling and code clarity. Still very good), so it’s not difficult to get started.
- Email: [email protected]