Get In Touch With Me!
I've always been passionate about personal
growth and development and I'm always looking for ways to improve. I'm
interested in self-improvement and a firm believer that anyone
can change their life through knowledge and focusing on themselves.
I'm dedicated to school and learning and embark on personal projects
to deepen my understanding of software development.
Through my classes at school and self-driven learning I strive
to improve my skills and advance my knowledge.
-Tyler Kiesman
I was responsible for designing and prototyping a new UI to replace the current display for a content management system. I developed a bubble navigation system that displayed content property information and used nodes to convey the associations between different pieces of content in the system. I also researched ways to transform some of the information which was previously being displayed in tables into charts and maps. Some examples included using box plots, choropleth maps, and chord diagrams to display this data. All of this was done using React and utilized the libraries D3.js and Rete.js for page manipulation and node creation.
I worked with IT communications and my engineering team to develop a Microsoft Teams bot that would take information from chats and display them on Cisco Webex boards that were located around the offices of the company. The back end was a Node.js server and used Amazon Web Services to help with storage, routing, and hosting the bot. I was also responsible for developing a computer vision model that would detect different bottles and determine the amount of liquid in them.
Developed a system for centralizing data from multiple programs and services that the company had developed. I worked to format all the data into a single JSON format, made it flow through a tool called NiFi, and stored it into a Solr database where it could be viewed and filtered using a platform called Banana. I also worked to replicate the same system this time using one piece of technology called Fluentd.
Worked on multiple assignments under the team responsible for the Service Line Analytics product. The first being a SQL script which would generate a report for the business analysts on the team to see what information was stored in one of their databases and what information was missing. The next was a Java program which would extract information from an Excel workbook and would generate a .csv file and a corresponding database migration which would insert the generated information into a database. Before, all the information had been put in by hand but this program turned what would take hours into a process that would take seconds. Lastly, I developed a generic algorithm for detecting duplications found in files that the team was receiving from entities sending them data. This algorithm was then implemented into their pipeline for receiving files.
| Title | Description |
|---|---|
| Seneca Park Zoo Society Business Intelligence Dashboard | Acted as the team lead on a project where we worked with the Seneca Park Zoo Society to develop a business intelligence dashboard. The dashboard collected data from a variety of resource streams and consolidated them into an easy-to-navigate web app. This required my team to develop a project plan which defined our process methodology, goals and scope, deliverables, risk management, scheduling, and metrics. We followed a Scrum approach, having two-week sprints, creating user stories, estimating tasks, calculating our velocity, and conducting retrospectives. We worked to gather requirements and develop an MVP for the zoo. The project was made with NodeJS and had a React frontend along with an Express backend to serve API calls. |
| Number Guessing Game | Although I wouldn't consider this a major project by any means, this small exercise introduced me to JavaScript and jQuery. It's a basic guess the number game that I created to get a better feel for how JavaScript interacts with a webpage and to get used to the syntax of jQuery. When a user guesses a number the webpage responds by giving hints to the user until they guess the correct number. A link to this game can be found here. |
| DressUp | Took part in an online hackathon that Facebook hosted where the goal was to develop an application using their machine learning framework Pytorch. I developed a basic Android application where a user could upload pictures of the clothing in their wardrobe and the application would build a stylish outfit based on color, clothing trends, and the weather. I developed a Faster RCNN Object Detection model in Pytorch and used that along with a K-Means clustering algorithm to help determine clothing types and the color of the clothes. These would then be combined to build an algorithm that could build stylish outfits that were weather appropriate. |
Get In Touch With Me!