# Ai Learning Roadmap

## Stage 1 — Ai literacy
Learn Artificial Intelligence, algorithms, models, data, automation and responsible use.

## Stage 2 — Data foundations
Learn datasets, features, labels, cleaning, structured and unstructured data, privacy and data quality.

## Stage 3 — Programming foundations
Learn Python, functions, debugging, testing, Git, APIs and notebooks.

## Stage 4 — Machine Learning map
Compare supervised, unsupervised, semi-supervised, self-supervised and reinforcement learning.

## Stage 5 — Core problem types
Practise classification, regression, clustering, dimensionality reduction, anomaly detection, forecasting and recommendation.

## Stage 6 — Deep Learning and Generative Ai
Learn neural networks, CNNs, sequence models, Transformers, embeddings, LLMs, prompting and RAG.

## Stage 7 — Application building
Connect a model or API with a frontend, backend, database, authentication and human-review workflow.

## Stage 8 — Evaluation
Choose appropriate metrics, compare a baseline, analyse errors, test robustness and document limitations.

## Stage 9 — Production
Learn deployment, model serving, monitoring, drift, retraining, latency, reliability and cost.

## Stage 10 — Portfolio evidence
Publish the problem, design, implementation, evaluation, responsible-use review and lessons learned.

### Progress check
- [ ] I can explain the Ai map.
- [ ] I can frame a problem before choosing a tool.
- [ ] I can select and justify a baseline.
- [ ] I can evaluate more than one metric.
- [ ] I can explain risks and limitations.
- [ ] I can show one complete project.