Skip to main content

Challenge instructions

Instruction and application
In Progress

By the end of the hackathon, each team will deliver a 5-minute presentation, followed by3 minutes of questions, covering:

  • Your AI/ML solution: The core proposal.
  • Technical choices: Key data, methodology, and modeling.
  • Guardrails: Governance, ethics, and stakeholder considerations.
  • The future: Clear recommendations for next steps.

Success criteria​

  • A solution that is clear, coherent, and realistic for the problem you framed.
  • Explicit assumptions, constraints, and trade-offs, clearly explained.
  • Evidence that feedback has been used to refine your approach.
  • A presentation that prioritises clarity and impact, rather than maximum detail.Today, you'll work on framing your ML problem, defining your solution scope, and creating a first-pass technical design.

Hackathon Resource Pack​

If you haven’t already, or if you need to re-download it, download the Hackathon Resource Pack now. It contains:

  • Descriptions of the two scenarios.
  • Datasets for each scenario.
  • Presentation template to capture your thinking throughout the hackathon.

You’ll continue using this pack across both workshops 1 and 2.

AIMLF Hackathon Resource Pack.zipOnce downloaded, open the files and keep them accessible for the rest of the hackathon workshop.

Instructions​

Frame the ML problem​

Using the scenario description and what you observe in thedataset, translate the business problem into an ML problem in your presentation template:

  • Business objective: Clearly state the business goal you are trying to achieve.
  • Target variable (y): Identify what you would aim to predict or estimate.
  • Modelling type: Decide whether this is best framed as classification, regression, forecasting, or another ML task.
  • Potential features (X): What specific columns or data signals from the dataset will the model look at to make that prediction?
  • Success metric: Define how success would be measured and how it links back to the business need.

You can review the Module 1 content on framing business problems for ML solutions to support your thinking in this step.

Define solution direction and scope​

In this step, your goal is to agree on a clear, shared direction before moving into technical thinking. As a team, agree on:

  • The AI/ML approach you are proposing at a high level and how itaddresses the business problem.
  • What is in scope andout of scope for today.
  • The** business success criteria** you will use to judge your solution (e.g., reducing customer churn by 5%, saving 10 hours of manual labour per week). How does your technical metric above move this business needle?
  • Any key assumptions or constraints shaping your design.

If you need a refresher, review the Module 2 content on translating business needs into technical specifications to help clarify scope, assumptions, and success criteria for your solution.

Initial data, modelling, and evaluation thinking​

With a solution direction agreed, begin first-pass technical design. As a team, sketch out:

  • Data: What data you are using, where it comes from, and its limitations.
  • Pre-processing: Required transformations and your strongest predicted feature signals.
  • Modelling: Potential model options and why they fit your constraints.
  • Evaluation: How you will measure and assess model performance.

You can review the following learning content to support your work for this step:

Next steps in the hackathon​

In the next hackathon workshop, you’ll build on this foundation by refining your data and modelling decisions, defining how success will be evaluated, and considering governance, ethical, and practical implications.