Hello!

My name is Daniel Ocheltree and I am a senior at Virginia Tech majoring in Computer Science. I am interested in full stack development, but I favor front-end JavaScript work. I have working knowledge in JavaScript, HTML, CSS, Python, and Java. I have developing knowledge in C, PHP, MATLAB, R, and SASS. I am experienced with Angular 5, AngularJS, jQuery, MongoDB, NGINX, Google Maps API, Google Calendar API, AWS EC2, and AWS S3.


Projects

Kelly Pool Ball Generator

Created a non-overlapping number generator to assign balls for a game of Kelly Pool. Users can select the number of players and hide or show assigned ball numbers. All graphics were made with CSS.
Kelly Pool Generator




IEEE at VT Website

Created a new website for the IEEE Virginia Tech Chapter.
www.ieee.vt.edu




Simple Test Tool

The Simple Test Tool is an internal testing tool used at Capital One. It is an Angular 5 based web app and is hosted on an EC2 instance.
The test tool is used by an operations team to test account messages for Capital One's EASE web and EASE mobile platforms.




ClusterUI

"ClusterUI is the premier comprehensive solution to convert days, weeks, or years of video data into easily-findable references which we call Tracks.
From those Tracks, we automatically group similar Clusters of Tracks to provide tools like filtering by media type, location, tag, and date, along with sorting by ethnicity, gender, and resemblance."


The ClusterUI is an AngularJS web app displaying data from a Python server. I worked on this product with a small team of three people.

Some of my responsibilities were:





Intelligent Solutions Group Website

I created a marketing website for the Intelligent Solutions Group (ISG) from scratch with AngularJS and PHP forms. This project was done mostly independently. While I had input on key design decisions, I worked with a team to create the general layout and content.

www.intelligentsolutionsgroup.com is not currently live.





Facial Rotation Annotator

Created a web app from scratch with JavaScript, HTML, CSS, and Python to calculate the roll, pitch, yaw, size, and location of faces in images using Euler angles and rotation matrices. Used by annotators to gather facial rotation data for Progeny’s SPOTR recognition software. The app is to be used by annotators to gather facial rotation data for Progeny’s SPOTR recognition software. The user loads an individual picture with a URL, file path, or .json file with a list of pictures. They then click 9 points indicated on the diagram (see image 3) and a face mesh is projected on the picture. The roll, pitch, yaw, x and y pixel location, and scale are displayed. The user can alter the roll, pitch, and yaw as well as the face mesh's location and scale with mouse controls or buttons, as they see fit. Once the mesh is in place, the user presses the save button and the data is written to a .txt file. If a JSON file with multiple pictures was loaded, the next picture is displayed. The web app is launched with a bash script that starts up the Python back-end and the JavaScript/HTML front-end.





Ship Data Mapper

I created a web app from scratch with JavaScript, HTML, CSS and the Google Maps API that maps out the locations of a GeoJSON with clickable points. The user runs a Python script that traverses through a MongoDB server full of ship data and puts it into a GeoJSON. The user then loads the GeoJSON using the load button, or drag and drops the file into the window. Ship locations are mapped out and clickable points are created. When the points are clicked, all known data about the ship is displayed. The satellite images, the ship types, and timestamps for the pictures are displayed.





Notice to Mariners

I wrote a Python script to pick up keywords from Notice to Mariner (NtM) documents to speed up the process of logging landmarks and geographic features. The script outputs the data as a JSON sorted by frequency and a bar graph of the top 25 words that are used. In order to run the script, the user must feed it a MongoDB server with the NtM data. The algorithm is optimized so it can run efficiently when given tens of thousands of inputs.
I also created HTML forms for the National Geospatial-Intelligence Agency (NGIA). The forms are used by mariners across the United States to report new landmarks, changes in geographic features, and buoy locations.




Camera Data Mapper

I created a web app from scratch using JavaScript, HTML, and CSS that displays a map of a facility and its cameras locations. The user can add camera locations to the map and link them to a Python web server. The web app displays when cameras detect a person. Users can click the cameras to display video. The camera data can also be filtered by date and time. This served as a proof of concept for an alternate version of the ClusterUI.



Contact

Email
Resume
LinkedIn
CodePen
GitHub