Applying Phase 2 skills
Instruction and application
In ProgressThis hackathon focuses on four key skills that you will practise by working through an end-to-end AI/ML solution:
- Skill 1: Model development and training
Implementing and training ML models using Python and structured datasets, applying appropriate preprocessing, feature selection, and baseline modelling approaches.
- Skill 2: Model evaluation and performance improvement
Evaluating model performance using appropriate metrics, comparing approaches, and improving results through iterative experimentation and refinement.
- Skill 3: Interpreting model behaviour and limitations
Analysing predictions, errors, and model behaviour to understand where models perform well or struggle, and linking those outcomes to business impact.
- Skill 4: Practical risk and governance awareness
Identifying practical risks and limitations based on model outputs, considering the implications of incorrect predictions, and proposing realistic safeguards and monitoring approaches.
Applying skills to real-life scenarios
Building on the solution design work completed in Hackathon 1, this hackathon focuses on the next stage of the AI/ML lifecycle: testing ideas in practice and learning from model behaviour and performance.
The scenarios reflect the kinds of challenges organisations face, where teams must evaluate how well models perform, compare approaches, and make informed decisions about which solutions are reliable, practical, and suitable for real-world use.
Action item: Discussion
**Directions:**Share your thoughts in the chat or come off mute
- What makes a machine learning model "good enough" to use in practice?
- If two models perform differently, what factors besides accuracy might influence which one you choose?
- Why can testing and improving a model iteratively be more valuable than trying to get everything right the first time?