Khoury master’s students place first in Harvard agriculture datathon
Author: Juliana George
Date: 01.23.25
![Left to right: Rakshak Kunchum, Laasya Anantha Prasad, and Krishna Venkatesh](https://s44427.pcdn.co/wp-content/uploads/2025/01/Team-pic800x533.jpg)
When they decided to enter the Harvard Data Science Initiative’s first-ever Agri Datathon, Khoury College graduate students Rakshak Kunchum, Laasya Anantha Prasad, and Krishna Venkatesh weren’t aiming to win the entire competition.
Although each of the three team members, who called themselves CodeCultivators, had previously participated in 24-hour hackathons, the Agri Datathon was a different beast — a three-day marathon focused on agriculture, a topic that no one in the group had experience with. But that didn’t stop them from earning first place and a $900 cash prize for their time series data project evaluating the economic implications of an aging farmer workforce in the US.
Prasad and Venkatesh first came across a flyer for the event while browsing LinkedIn and decided to bring in Kunchum, a former classmate. All three were experienced coders accustomed to working on a deadline, but nothing could have prepared them for the grueling hours of the datathon.
“We worked for almost the entire day from morning until night trying to finish our submission,” Kunchum said. “It was very challenging and very intense.”
Participating teams assembled in Harvard University’s Science and Engineering Complex on the morning of October 4 for the datathon, which was created to get Boston-area students to engage with real-world agricultural problems normally reserved for land-grant universities. Each team selected one of three prompts based on data from the US Department of Agriculture’s National Agricultural Statistics Service, Harvard’s collaborator in planning the event. CodeCultivators chose a prompt that asked participants to analyze the relationship between farmer age and agricultural production on a state level in the US.
![About 30 participants in the Agri Datathon post for a photo on the event stage](https://s44427.pcdn.co/wp-content/uploads/2025/01/Datathon-group800x350.jpg)
To evaluate the substantial data, which spanned from 1997 to 2022 and touched on agricultural sales, land use, and state-level farmer demographics, CodeCultivators first had to merge it into a single dataset. From there, they identified state and regional trends in farmer age with a heatmap that helped them locate the states where farmer age had spiked.
“We had an interactive plot made where the users could hover over each state and look at the median age of farmers, look at what the common crop yield looks like, and look at them over the years,” Prasad said.
Then they built a seasonal autoregressive integrated moving average model, a time series forecasting model that allowed them to predict the average farmer’s age in different states for the next five to 10 years. Using the model, the team discovered an upward trend in the average age of farmers across the US, which, when measured against sales data on products like rice, corn, and cotton, suggests that production may fall if the average age continues to climb.
“We figured out that the average age of farmers is increasing, and this is probably because the technological advancements in these fields are also increasing. So older farmers can put in less physical effort and still get the same produce that they did about 10 or 15 years ago,” Prasad explained. “At the same time, we saw that the younger generation is taking a step back and is a little hesitant to get into this field.”
Less than a month after the team submitted the fruits of the comprehensive three-day project, which included a video presentation to the judges and an academic paper, they learned they had earned the top prize.
The Khoury Network: Be in the know
Subscribe now to our monthly newsletter for the latest stories and achievements of our students and faculty