News Search

ACC/A2’s Summer Data Cohort Program

  • Published
  • By Tech. Sgt. Emili Koonce
  • Air Combat Command Public Affairs

This summer the Air Combat Command campus was a buzz with 13 members participating in the ACC/A2 summer data cohort program at Joint Base Langley-Eustis, June 6 - August 26.

Cohort members included a combination of data scientists, Department of Defense Cyber Scholarship recipients, PALACE Acquire interns and a cadet from the U.S. Air Force Academy.

Before arriving at JB Langley-Eustis' cohort members were assigned a senior mentor from ACC/A29, A2’s intelligence division. The senior mentor developed a project focused on a research theme that ties into their expertise such as targeting, cyber, analysis or data engineering.

“We always try to tie the project to something operational,” said Adrian Peter, ACC/A29 senior data scientist. “I work with cyber units within ACC so am intimately aware of some of the problems facing this community so, I wanted the cohorts to create a tool to assist defensive cyber analysts.”

Andres Davila-Corujo, Irving Santana-Vega and Sarah Osborn were assigned to project Delta Seeker. They were selected for Peter’s project because of their strengths and expertise and once matched with their mentor and project, the cohorts were ready to dive into problem solving. 

The primary objective for the cohort was to tackle their projects and learn from their mentors. Additionally, ACC/A2 strove to ensure the cohort left with increased technical skills and hands-on experience in the project development life cycle. They also gained a stronger understanding for A29’s role in analysis as well as application of intelligence, surveillance and reconnaissance data.

Cadet Ria Singh, a U.S. Air Force Academy Cadet in their senior year, engineered data for the project.

“In this particular project (Delta Seeker) we split it into two components, data engineering and analytics,” said Peter. “Cadet Singh, along with the team, took on one of the project’s biggest challenges, managing a large volume of data.”

In total the members assigned to this project were managing and analyzing 40 gigabytes of data and 140 million records. This was just for development and prototype purposes, but Peter has hope the program can be worked for cyber analysis mission needs.

“To ensure our adversaries are not on our network or to defend against a malicious attack, we need to sift through millions and even billions of records,” said Peter. “To have the ability to do so in real time we employ automation.”

In preparing sets of data or records to be used in automation it is necessary to first clean up the data. This is also known as data scrubbing, but no matter the name, it is tedious and precise work to ensure there is no incorrect, incomplete, duplicate or otherwise erroneous data within the records.

“After the data is cleansed, I ensure the proper metadata is applied so there is a governance on who can receive the data as well as information available for who wrote it, where it came from and other important metadata,” said Singh. “After this step I ensure the data sets reach an elastic search index within a database that the Delta Seeker program will use to run its analyses.”

After Singh applies the metadata and indexes the data set, the data is ready to run through the analysis program Davila-Corujo, Santana-Vega and Osborn have been working on in tangent with Singh.

This team effort had its biggest eureka moment in early August according to Osborn.

“I would say our proudest success was when all the components came together,” said Osborn, DoD Cyber Scholarship graduate. “The plugin generated correctly, you could interact with and filter the visualizations, and you could connect to the API (application programing interface) to train ML (machine learning) models and quickly visualize the results.

“The team had to overcome a lot to get to that point, so to see everything working together was an immense relief and accomplishment.”

To conclude the summer program, teams prepared an overview of their project to include their accomplishments, challenges and reflections on the program.

Interested ACC/A2 Airmen filled the top floor of the small presentation room to hear the cohort’s new ideas and solutions. Additional attendees included peers, mentors, senior mentors along with ACC/A2 leadership.

For Osborn this was not her first final presentation. She has attended this summer program since 2019. However, this is her first in-person summer program.

"I feel I've learned so much over the past three summers, and I've met so many great people as a result,” said Osborn. “The experience so far has been invaluable.”

Her time with ACC/A2 is not over as she will continue to work within ACC/A29 as a computer scientist.

"I'm looking forward to contributing to projects that have real-world impact,” said Osborn. “That was the pull of working in the public sector for me, and why I applied to the Cyber Scholars program; I wanted the chance to work on projects with very real and immediate significance to something greater than me.”

If you’re interested in attending ACC/A2’s summer data cohort program next year, work with your chain of command to submit an application to ACC/A2’s data operations and data lab branch,