The Sherubtse Hackathon 2026 is a department-level event in which student teams solve a real-world data problem within a fixed time window using data science, machine learning, or data analytics tools and techniques. The event is meant to encourage teamwork, applied problem-solving, and creative use of data. What is more, it aims to connect classroom learning with real-world problems. 

Sherubtse Hackathon 2026 Challenges

E-Commerce

Analyse consumer behaviour and predict e-commerce outcomes in Bhutan. Explore purchasing patterns, identify factors associated with customer behaviour, and build a model to predict a relevant outcome such as purchase likelihood, spending level, or customer segment.

Tourism

Analyse visitor behaviour and predict tourism outcomes in Bhutan. Explore visitor characteristics and travel patterns, identify factors influencing tourism outcomes, and build a model to predict an outcome such as visitor satisfaction, spending, or length of stay.

Agriculture

Analyse agricultural factors and predict crop production in Bhutan. Explore relationships between environmental, farm, and management factors and crop production, identify important factors, and build a model to predict crop yield or production.

Digital Payments

Analyse digital payment behaviour and predict payment-service usage in Bhutan. Explore patterns of adoption and usage, identify factors associated with digital payment behaviour, and build a model to predict an outcome such as adoption, usage frequency, or transaction category.

Education

Analyse student behaviour and predict educational outcomes. Explore factors related to student performance and engagement, identify important patterns, and build a model to predict an outcome such as academic performance, pass/fail status, or engagement level.

Healthcare Services

Analyse healthcare service utilisation and patient experience. Explore patterns in service use and patient experience, identify factors associated with these outcomes, and build a model to predict an outcome such as satisfaction, service utilisation, or likelihood of returning for services.

Challenge

Use the provided dataset to investigate a real-world problem within your chosen topic. Participants must:

1. Explore the data and identify meaningful patterns or trends.
2. Analyse relationships between relevant variables and the selected outcome.
3. Build and evaluate at least one machine-learning model to predict the selected outcome.
4. Interpret the results and identify important factors associated with the outcome.
5. Communicate the findings through clear visualisations and a concise data-driven story.

Dataset and Data Usage

Synthetic datasets have been prepared for each challenge topic to provide a consistent and fair basis for the hackathon. The datasets represent plausible real-world scenarios and are designed to support meaningful data analysis and machine-learning tasks.

• Teams must use the provided dataset for their selected topic.
• Basic preprocessing, such as handling minor missing values, encoding categorical variables, scaling features, and feature engineering, is permitted. Extensive data cleaning is not expected.
• External datasets should not be used for the main analysis.
• Teams should identify relevant assumptions and limitations of their analysis and model.
• Findings should be interpreted within the context of the provided synthetic data and should not be presented as factual claims about the actual population or sector in Bhutan.

• Judging Criteria
  The following are the assessment criteria for the hackathon: a. Problem understanding (15%), b. Technical approach (30%), c. Results and evaluation (20%), d. Creativity/innovation (15%), and e. Presentation and clarity (20%)