DevOps Engineer and Developer in Mueang Chiang Rai Chiang Rai, Thailand

I built a web application in Python 3 and Django to provide a visual representation of stock exchange data from all world-major markets. The app gives the ability to present and visualize data with JavaScript graphs. It is also possible to process data using Jupyter Notebooks equipped with Python 3 libraries (like Pandas) and a database connection.
As one of the 12 team members, I was responsible for delivering the build, test, and deployment pipeline based on Atlassian Bitbucket and Bamboo (one master was distributing jobs for up to 25 agents residing on Spot instances). AWS ECS served the web application and all required services. The configuration was kept as a CloudFormation stack, which also deployed all the required services, including Jenkins with a Groovy-based configuration and PostgreSQL database for Atlassian solutions (including Jira, Bamboo, and Bitbucket). I was documenting solutions with Sphinx documentation framework.
The project was consulted with and utilized by academics, both students and lecturers. It also had the further potential of being used by professional services. In its peak time, there were 25 to 50 users.

On this project in Nokia Networks, I handled multiple duties and roles.
The primary one was integrating C/C++ low-level components to build firmware that needed to run custom Linux on Base Transceiver Station hardware. Using Jenkins running on Red Hat, I automated code polling from Git and SVN servers, wrote Make and CMake scripts, and executed tests on remote Windows machines. These Windows machines were equipped with in-house testing software that was directly connected to BTS hardware. I also created artifacts and informed the owner if anything went wrong during the process. I was documenting every new solution with Confluence pages and making sure it was clear and understandable to co-workers.
In a secondary role, I supported the radio frequency development team of over 50 members, chiefly dealing with pipeline maintenance, responding to developer tickets about new development/test environment creation, and working as a member of one of the seven integration teams.

Webiny is a CMS built by a small startup company around AWS serverless technologies like Lambda, CloudFront, API Gateway, or S3. It uses React as a web application framework. The product is in its early stage and maturing quickly, so it will receive production-grade status soon.
As a new technologies enthusiast, I spotted huge potential in this solution in terms of decreasing costs and scalability, thus, I started experimenting with it using my own AWS account.
Right now I'm actively contributing to improving the administrators' and end-users' experience. I'm answering questions in the community forum, helping to do the first deployment, and explaining how the back end works. I'm also delivering security improvements, reporting bugs, and providing ideas about how to make Webiny a high-class product.
My descriptions are equipped with screenshots and GIFs or short videos to make the problem easy to understand. I'm also getting feedback, so I can improve my documents.
I'm a recognized contributor on Webiny's GitHub repository page, Github.com/webiny/webiny-js, as Piotr Palka. The community forum is available under the link below.

As an AWS engineer focused on cost and scalability I analyzed the current AWS infrastructure architecture to identify optimization issues and prepare a plan to fix them. One of them was to find out the access frequency to the Node.js server EC2 instance is very low so it's worth moving it to ECS Fargate containerization solution to avoid idle costs generated by the instance that is running but not processing anything.
AWS networking cost improvements optimization, security tightening, and documentation were also involved which allowed more traffic being handled by AWS internal network. The simple but descriptive documents produced at the end of the project, will allow new developers to dive into the code quickly and without employing costly and time-consuming reverse engineering.
By moving resources from fixed instances to the dynamical scale model and improving networking I was able to cut the operational costs by half when the system is not heavily utilized. The auto-scaling built on AWS Lambda and AWS ECS can process every clients' request separately which improved reliability and repetitivity of the whole process with the general benefits to both the platform owner and his customers.