My name is Yesh Nagelli and I am a student researcher interested in computer science and machine learning.
Many problems persist in education today such as inequity, poor student outcomes, and lack of proper resources in many communities. I use computer science and machine learning to understand student outcomes, expand access to computing, and build tools that support educators.
Founder & Chief Executive Officer
Growing up, I was fascinated by technology and computer science, but felt overwhelmed learning how to code on my own. A formal computer science class was not offered to me until my freshman year of high school. From that class, I learned so much about computer science, and had so many opportunities to collaborate with others and create things we were passionate about. I wanted everyone to have access to the same opportunity I did, especially younger students who lacked structured STEM programs in their schools.
I started Commit2Code to bridge the access gap in computer science. I developed a 12 week curriculum that teaches students to program in Python through short lectures, games, and projects. I have partnered with organizations such as the Boys and Girls Club of Joliet, Communities in Schools of Chicago, and the Plainfield Area Public Library to reach underserved students.
fig. 1 — Commit2Code, by cohort
50+ students reached, and counting.
Independent research
Model deployed · paper submitted for publication
Throughout many of the classes I taught through Commit2Code, I noticed students began to stop attending or have irregular attendance. I started to consider the data and was fascinated by the similar characteristics of those who stopped attending. It was clear that there were some predictors of dropout.
I began to think about how dropping out of high school was common in many school districts near me and in some that I work with through Commit2Code. After talking with many teachers and administrators about the issue, I understood that it was too difficult for teachers to continuously monitor thousands of students to determine if they are at risk of dropping out.
I realized that data science and machine learning could be used to develop a model that can predict a student's likelihood of dropping out. Students at risk would clearly be flagged for administrators allowing them to extend additional support.
I used data from the High School Longitudinal Study of 2009 from the National Center of Education Statistics to train and test the model. Furthermore, I developed a 5-page research paper outlining my methodology in developing the model.
The model is currently deployed and free to use for teachers and administrators.
After deploying DropCast, I was left with more questions than I started with. I started to consider how educators could act on probability, how certain variables can build inequality into the model, and how the model could inform educator intervention. I want to learn more about machine learning and computer science to learn how models can provide useful and responsible insights. I want to continue to research how machine learning can further support educators and develop DropCast accordingly.
I was selected into the BU Rise data science practicum program. Under this program, I researched machine learning methods for improving cross-region North Atlantic right whale upcall detection. I presented at the BU RISE symposium and devised a 5 page paper that was accepted into the IEEE MIT Undegraduate Research Technology Conference(URTC)
President
As president of my school's computer science club, I organized weekly coding competitions and tutorials to engage 50+ members in computer science. Tutorials I led included machine learning, APIs, and facial recognition. Furthermore, I launched team builds including an arcade machine and automated attendance system strengthing collaboration
Competitor
I have been a competitor on my school's math team since freshman year. I placed regionally and competed at state 3 consecutive years.
Founder
As someone who loved technology and valued educational equity, I was always frustrated by the fact that my school lacked a competitive robotics team. Since my sophomore year of high school, I worked closely with a club sponsor and school administration to start a team. I helped secure $2800 in funding through company sponsorships and prize money awarded in local competitions. Furthermore, I recruited underclassmen students and trained them across responsibilities to ensure long term program continuity .