About the role
You will join the Applied AI team and work on prediction problems that come from client briefs: forecasting demand, flagging risk, ranking and classifying. You own each model from raw data to a working API, and you explain your results to a non-technical client the way a real ML engineer does.
Roles and responsibilities
- Clean, explore and document real-world datasets before any modelling
- Build baseline models first, then improve them with evidence, not guesswork
- Choose the right metric for the business problem and report it honestly
- Engineer features and build reproducible scikit-learn / PyTorch pipelines
- Serve trained models through a FastAPI endpoint with input validation
- Write model cards that explain limitations, bias and intended use
- Ship every deliverable to GitHub with a README a client could follow
By the end, you will be able to
- Take a vague business question and frame it as an ML problem
- Train, tune and compare classical and deep-learning models
- Detect overfitting, data leakage and class imbalance before they bite
- Deploy a model behind an API and monitor it for drift
- Present results with charts and plain-language trade-offs
Questions you will be able to answer
How do food-delivery apps estimate your delivery time before the rider even moves?
Why can a model with 99% accuracy still be useless?
How does a bank decide in seconds whether a transaction looks like fraud?
What actually happens inside a neural network when it "learns"?
What you might build
- A delivery-delay predictor served as an API
- A churn model with an explainability report
- An image or text classifier using transfer learning
Every intern gets different tickets. These are examples of the kind of work, not a fixed list.
How the first three months progress
Week 1
Onboarding & Python for ML
dev environment, Git, virtualenv, NumPy, pandas basics, Jupyter workflow, project README & repo hygiene
Week 4
Supervised learning: classification
logistic regression, trees, precision/recall/F1, confusion matrix, class imbalance
Week 8
Neural networks
PyTorch/Keras basics, MLP training loop, overfitting & regularisation, learning curves
Week 12
Capstone delivery
end-to-end ML product, demo & documentation, model card, final presentation
Four- and six-month internships add advanced specialisation, a major client project, open-source work and interview preparation.
Tools you will use
Who should apply
- Comfortable writing basic Python (loops, functions, lists, dictionaries)
- Curious about how data turns into decisions
- Students in any year, recent graduates and career switchers are welcome