Code Safari
All subjects

The ML guides & blogs

ML

Train, evaluate, ship, and monitor machine learning models — and build the data plumbing they depend on.

10 guides & blogs·5 stages·≈ 2 hr·6 more being written

Start with Common Foundations — it counts on every subject.

  1. 01

    Machine learning foundations

    ≈ 1 hr

    What "learning from data" really means — the three flavours of ML, overfitting, and why data quality decides everything.

    Machine Learning for Curious Minds (No Math Required)A plain-English guide to how machine learning actually works — learning from data, the three flavors of learning, overfitting, why data quality decides everything, how neural networks learn, and how a model goes from idea to deployment. No math required.6 chapters·≈ 1 hr·also under AI
  2. 02

    Neural networks

    42 min

    From a single neuron’s weights and activations to training end to end, and the CNN/RNN/Transformer family tree.

    Neural Networks Basics (No Math Required)A plain-English guide to neural networks — what they actually are, the anatomy of a neuron (weights, biases, activation functions), training one end to end from random dials to reading digits, and the family tree of architectures: CNNs, RNNs, and the Transformer behind ChatGPT.4 chapters·42 min·also under AI
  3. 03

    Data engineering for ML

    Models are downstream of data — the pipelines, features, and labels that feed them.

    Data pipelines & feature engineeringbeing written

    Will cover: batch vs streaming pipelines · feature stores · labelling strategies & label quality · train/serve skew

    Working with datasetsbeing written

    Will cover: dataset versioning · class imbalance & sampling · data leakage & how to spot it · privacy & PII handling

  4. 04

    Training at scale

    From a notebook experiment to a reproducible training run you can trust and repeat.

    Experiments & reproducibilitybeing written

    Will cover: experiment tracking · hyperparameter search · seeds, determinism & environment pinning · GPUs & distributed training basics

    Evaluation, honestlybeing written

    Will cover: choosing the right metric · validation strategy & test-set hygiene · baselines before models · error analysis

  5. 05

    MLOps: serving & monitoring

    The model is 10% of the system — deployment, latency, drift, and retraining loops are the job.

    Model servingbeing written

    Will cover: batch vs real-time inference · model packaging & registries · latency, throughput & hardware · A/B testing & shadow deployments

    Monitoring & driftbeing written

    Will cover: data drift vs concept drift · monitoring predictions in production · feedback loops & retraining triggers · when to roll back a model

Shelves don't teach. Guides do.

Start: Machine Learning for Curious Minds