Databricks AI Social Impact (DAISI) Challenge - Singapore

Overview

The Databricks AI Social Impact (DAISI) Challenge - Singapore is a national student competition designed for students from Singapore's Institutes of Higher Learning (IHLs). Participants are tasked with tackling real societal problems by leveraging Singapore's national open data. The entire solution must be built on the Databricks platform, utilizing its free edition, and teams will benefit from mentorship provided by Databricks engineers throughout the development process. The challenge culminates in pitches to Databricks APJ leadership. This challenge provides nine distinct problem statements organized across three main tracks: Healthy Nation, Green & Resilient City, and Liveable & Inclusive City. Each problem statement is grounded in readily available open government datasets from sources such as data.gov.sg, SingStat, and other agency data, ensuring no cost or login is required to access the data. Teams are expected to develop an end-to-end solution, encompassing data ingestion, data engineering, machine learning components, and a live dashboard demonstration. Participation in the DAISI Challenge offers several benefits, including the opportunity to work on significant societal issues like ageing, climate resilience, housing, mobility, and community. Students will gain hands-on experience with the Databricks platform, a tool widely used by thousands of companies globally, at no personal cost. Furthermore, each team will be assigned a Databricks engineer as a mentor for a two-week build sprint. The competition features a prize pool valued at up to US$24,500, with the top-performing team earning the opportunity to present their solution at the launch of the new Databricks Singapore office. All teams will develop their projects using Databricks Free Edition, which eliminates any cost barriers, credit card requirements, or infrastructure management. This free edition includes access to notebooks, SQL warehouses, dashboards, Genie, Lakeflow, Unity Catalog, and MLflow, providing a comprehensive environment for building data and AI solutions. The challenge emphasizes practical, runnable pipelines rather than theoretical concepts, encouraging participants to develop functional tools that can generate genuine insights and be used by non-technical stakeholders.

What to build

Participants must build a working data and AI solution on Databricks Free Edition that addresses one of the nine specified problem statements across three tracks: Track A — Healthy Nation (caring for a super-aged Singapore), Track B — Liveable Nation (sustainability, mobility, and the urban environment), and Track C — Inclusive Nation (opportunity, affordability, and social mobility). The project must utilize Singapore national open data (e.g., Data.gov.sg, SingStat, LTA DataMall, NEA, HDB, URA) and be entirely built on the Databricks platform. Submissions should demonstrate a runnable data and AI pipeline, not just slideware, typically showcasing data engineering (ingestion and transformation into a clean, governed table), analytics or AI (a model, forecast, risk score, agent, or natural-language layer producing genuine insight), a usable output (dashboard, app, or interface for non-technical stakeholders), and clear social impact.

Prizes

Key dates

• Idea submission closes2026-10-06 23:59 SGT
• Deadline for final submission (shortlisted top 10 teams)2026-10-27
• Deadline2026-10-07 00:00 SGT

Eligibility

Participants must be of legal age of majority in their country of residence. Only college students are eligible; companies and professional organizations are excluded from participation. Only students from specific countries/territories are included.

How to enter

1. Form a team of 1 to 4 students from Singapore's Institutes of Higher Learning.
2. Choose one of the nine problem statements across the three tracks.
3. For Round 1, submit a 1-page concept note or 3-slide pitch deck (PDF) covering the chosen problem, proposed solution, datasets to be used, and intended Databricks architecture, along with team member details, by October 6, 2026, 11:59 PM SGT.
4. For shortlisted top 10 teams (Round 2), submit before Demo Day on October 27, 2026, a 3-minute maximum demo video (YouTube or Vimeo link), Databricks notebook/repo link, a written project description on Devpost, a pitch deck (up to 10 slides, PDF), and screenshots of the dashboard/app/Genie space, plus a short 'what's next' note. All submissions must be in English.

Submission requirements

• Build a working data and AI solution on Databricks Free Edition.
• Address one of the nine problem statements from the three tracks.
• Utilize Singapore national open data (e.g., Data.gov.sg, SingStat, LTA DataMall, NEA, HDB, URA).
• Solution must be built on the Databricks platform.
• Show a real, runnable pipeline, not just slideware.
• Demonstrate data engineering (ingestion, transformation into clean, governed table using Lakeflow, Delta, Unity Catalog).
• Include analytics or AI (model, forecast, risk score, agent, or natural-language layer producing genuine insight using MLflow, Foundation Model APIs, Genie).
• Provide a usable output (dashboard, app, or interface for non-technical stakeholders).
• Clearly explain who benefits and how (social impact).
• Teams must consist of 1 to 4 students.
• All submissions must be in English.
• Late submissions are not accepted.

Judging criteria

• Problem fit & social impact: Clearly addresses the chosen problem; real, describable benefit.
• Solution quality & originality: Thoughtful approach, not a generic dashboard.
• Data & technical feasibility: Named open datasets; credible Databricks architecture; buildable in two weeks.
• Clarity of submission: Concise, well-structured concept note.

Sponsors & partners

Resources

• Participant Guide