The Official Site Of NAIT (Native AI Teams)

Learn

Hands-on AI/ML curriculum — new lessons every day except Sunday, free.

Rules, Heuristics & Symbolic AI

  1. Ship the rule-based baseline first: why 20 lines of if/else gate every good ML project22/07/2026
  2. Rule learning: RIPPER, decision lists, and models that output human-readable rules30/07/2026
  3. RuleFit and skope-rules: mining crisp rules out of a trained ensemble04/08/2026
  4. Business rules engines: Drools, GoRules, and when domain logic must not be learned12/08/2026

Feature Engineering & the scikit-learn Toolkit

  1. Encoders shootout: one-hot, ordinal, target, and hashing — when each wins and when it leaks22/07/2026
  2. Scaling and normalization: StandardScaler vs MinMaxScaler vs RobustScaler, and the models that don't care30/07/2026
  3. Rolling windows and lag features: time series into supervised learning without leaking the future31/07/2026
  4. sklearn Pipeline and ColumnTransformer: preprocessing that cannot leak11/08/2026

Modern Data Engineering

  1. Parquet and Arrow: why columnar formats make your queries 100x faster21/07/2026
  2. DuckDB: an analytics warehouse that fits in your laptop's pocket25/07/2026
  3. Polars vs pandas: the DataFrame rewrite that changed the default31/07/2026
  4. ELT and dbt: transforming data where it lives, with tests and lineage11/08/2026

Multi-Model Inference & Serving

  1. From pickle to production: wrapping a model in a real API that won't fall over21/07/2026
  2. ONNX: one model format to run everywhere, from server to browser24/07/2026
  3. Dynamic batching: the single biggest throughput lever in model serving30/07/2026
  4. Serving many models on one GPU: Triton, model repositories, and hot-swapping10/08/2026

Classical Machine Learning

  1. Linear regression is not boring: assumptions, diagnostics, and when it beats deep learning20/07/2026
  2. Logistic regression as a classifier you can actually explain24/07/2026
  3. Decision trees: how they split, why they overfit, and how to read one30/07/2026
  4. Random forests: why averaging many bad trees makes one great model08/08/2026

Exploratory Data Analysis & Statistics

  1. First hour with a new dataset: a repeatable EDA checklist in pandas20/07/2026
  2. Missing data is a message: patterns, mechanisms, and honest imputation24/07/2026
  3. Outliers: detect, understand, and only then decide to delete30/07/2026
  4. Correlation is not causation, but it's also not nothing: reading correlation matrices right08/08/2026

Mathematics for Machine Learning

  1. Vectors, dot products, and cosine similarity: the geometry behind embeddings18/07/2026
  2. Matrix multiplication is all you need: shapes, batching, and why GPUs love it23/07/2026
  3. Eigenvalues and SVD without tears: compress an image to see them work30/07/2026
  4. Derivatives and gradients: the one calculus lesson ML actually uses07/08/2026

Neural Networks from the Ground Up

  1. A neuron is just a weighted sum: build a perceptron in 20 lines of NumPy18/07/2026
  2. Backpropagation demystified: hand-compute the gradients, then verify with autograd23/07/2026
  3. Activation functions in 2026: why ReLU won and what's replacing it30/07/2026
  4. Weight initialization: why your deep net trains or dies before step one07/08/2026

ML Foundations & Chatbots

  1. The bias-variance tradeoff, finally explained with runnable plots11/07/2026
  2. Data leakage: the silent killer of ML projects14/07/2026
  3. Feature stores and reproducible training pipelines30/07/2026
  4. Model monitoring in production: drift, decay, and alerting06/08/2026

Computer Vision

  1. Object detection today: from YOLO to open-vocabulary models11/07/2026
  2. Image segmentation with Segment Anything and its successors15/07/2026
  3. How diffusion models generate images, explained with a runnable toy17/07/2026
  4. Vision-language models: teaching an LLM to see06/08/2026

RAG & Agentic Development

  1. Build a minimal RAG pipeline: chunk, embed, retrieve, answer11/07/2026
  2. Chunking strategies that actually matter for retrieval quality13/07/2026
  3. Choosing a vector database in 2026: pgvector vs the specialists16/07/2026
  4. Hybrid search: combining BM25 keyword and dense vectors05/08/2026

LLMs & Transformers

  1. How transformer self-attention actually works, with a runnable toy11/07/2026
  2. Tokenization pitfalls: why your LLM miscounts characters and money12/07/2026
  3. Context windows, KV-cache, and why long prompts get expensive15/07/2026
  4. Fine-tuning vs LoRA vs prompting: choosing the cheapest path that works05/08/2026

Gradient Boosting with XGBoost

  1. Gradient boosting from the ground up: why XGBoost and LightGBM still rule tabular data in 202631/07/2026
  2. Tuning XGBoost without fooling yourself: early stopping, learning rate, and the parameters that matter31/07/2026
  3. Categorical features and missing values in XGBoost: the native path vs one-hot31/07/2026
  4. XGBoost vs LightGBM vs CatBoost: an honest benchmark on the same data and the same budget31/07/2026
  5. Learning to rank with XGBoost: pairwise and listwise objectives for search and recommendations31/07/2026
  6. XGBoost in production: GPU training, out-of-core data, model persistence, and fast inference31/07/2026
  7. XGBoost pitfalls: the mistakes that quietly wreck a boosted model31/07/2026
  8. Debugging a boosted model: learning curves, tree dumps, and plots that explain a prediction31/07/2026
  9. Squeezing more out of XGBoost: feature engineering, custom objectives, and constraints31/07/2026
  10. Reproducible XGBoost: seeds, pinned versions, and a validation protocol you can defend01/08/2026
  11. Keeping an XGBoost model honest: drift monitoring, champion/challenger, and retraining cadence03/08/2026
  12. Forecasting with XGBoost: turning a time series into a supervised problem without cheating04/08/2026

Distributed ML with PySpark

  1. Why PySpark for ML: when your data outgrows pandas and a single machine24/07/2026
  2. PySpark DataFrames, shuffles, and fast UDFs: pandas_udf + Arrow beat Python UDFs 10x24/07/2026
  3. Spark ML feature pipelines: VectorAssembler, StringIndexer, OneHotEncoder, and the Pipeline API24/07/2026
  4. Training models with Spark MLlib: LogisticRegression, RandomForest, and GBTClassifier24/07/2026
  5. Distributed hyperparameter tuning: CrossValidator, ParamGridBuilder, and TrainValidationSplit24/07/2026
  6. Clustering and recommendations at scale: KMeans and ALS collaborative filtering in Spark24/07/2026
  7. Text features in Spark: Tokenizer, HashingTF, IDF, and Word2Vec on large corpora24/07/2026
  8. Scaling single-node ML on Spark: pandas API on Spark, applyInPandas, and grouped training24/07/2026
  9. From Spark model to production: persistence, MLflow tracking, and batch scoring24/07/2026
  10. PySpark ML performance: partitions, data skew, caching, and the small-data trap24/07/2026

Channel Specials

  1. Best practices: avoid mutable default arguments02/12/2025
  2. Using timeit and perf to benchmark Python code04/12/2025
  3. Using FastAPI to build a mini REST API in Python08/12/2025
  4. Introduction to infographics and visual learning for Python/ML10/12/2025
  5. Introduction to benchmarking in Python16/12/2025
  6. Empirical: float precision errors in Python math24/12/2025
  7. Using matplotlib/plotly for infographic-style outputs28/12/2025

Data Science

  1. Best practices for robust feature pipelines01/12/2025
  2. Intro to model evaluation metrics02/12/2025
  3. Intro to model evaluation metrics02/12/2025
  4. Tools: Kubeflow, Vertex AI, MLflow Projects03/12/2025
  5. Empirical comparison of algorithms04/12/2025
  6. Tools: FastAPI, Docker, BentoML05/12/2025
  7. Expert: high-dimensional clustering06/12/2025
  8. Empirical: L1/L2 impact08/12/2025
  9. Tools: Feast, Hopsworks09/12/2025
  10. Expert: Bayesian optimization & Hyperband10/12/2025
  11. Intro to natural language processing11/12/2025
  12. Best practices for ML API design12/12/2025
  13. Tools: statsmodels, Prophet14/12/2025
  14. Intro to dimensionality reduction15/12/2025
  15. Tools: SHAP, LIME, InterpretML16/12/2025
  16. Best practices for ensemble tuning17/12/2025
  17. Empirical: algorithm benchmarks18/12/2025
  18. Tools: Evidently AI, WhyLabs19/12/2025
  19. Best practices for evaluating clusters20/12/2025
  20. Empirical: batch vs streaming stores23/12/2025
  21. Expert: distributed tuning with Ray Tune24/12/2025
  22. Tools: PyTorch, TensorFlow25/12/2025
  23. Intro to scaling ML inference26/12/2025
  24. Empirical: LSTM vs Prophet vs ARIMA27/12/2025
  25. Best practices for feature importance ranking28/12/2025

Data Engineering

  1. Tools: AWS Lake Formation, Glue Data Catalog01/12/2025
  2. Empirical: Airflow vs Prefect performance comparison02/12/2025
  3. Introduction to modern data warehouse design03/12/2025
  4. Introduction to ETL and ELT patterns04/12/2025
  5. Expert: real-time feature stores and ML stream inference05/12/2025
  6. Tools: Great Expectations, Soda Core, Deequ06/12/2025
  7. Introduction to streaming data architecture08/12/2025
  8. Best practices: domain ownership and federated governance09/12/2025
  9. Tools: Prometheus, Grafana, Airflow sensors10/12/2025
  10. Empirical: OLTP vs OLAP query performance comparison11/12/2025
  11. Introduction to stream processing concepts12/12/2025
  12. Expert: advanced lineage propagation across systems14/12/2025
  13. Empirical: Parquet vs ORC compression benchmarks15/12/2025
  14. Expert: event-driven orchestration with EventBridge and Step Functions16/12/2025
  15. Tools: dbt, Redshift Spectrum, Athena17/12/2025
  16. Best practices: schema evolution and metadata tracking18/12/2025
  17. Tools: AWS Athena Federation, Starburst, Trino23/12/2025
  18. Tools: AWS Athena Federation, Starburst, Trino23/12/2025
  19. Introduction to data pipeline monitoring and alerting24/12/2025
  20. Introduction to data pipeline monitoring and alerting24/12/2025
  21. Best practices: balancing read/write trade-offs25/12/2025
  22. Tools: Apache Beam, Flink, Dataflow26/12/2025
  23. Best practices: metadata enrichment and versioning27/12/2025
  24. Introduction to lakehouse architecture28/12/2025

More lessons

  1. Empirical: comparing SOLID vs non-SOLID Python code maintainability28/11/2025
  2. Best practices for wheel and source distributions29/11/2025
  3. Introduction to async/await in Python01/12/2025
  4. Expert: sustainable productivity systems for engineering teams01/12/2025
  5. Tools: black, ruff, pre-commit, mypy02/12/2025
  6. Best practices: responsible data handling and transparency03/12/2025
  7. Expert: async and GPU optimization patterns03/12/2025
  8. Best practices for reproducible, modular notebooks04/12/2025
  9. Introduction to technical teaching and mentorship05/12/2025
  10. Empirical: coupling and cohesion analysis05/12/2025
  11. Empirical: throughput comparison of streaming architectures06/12/2025
  12. Best practices: list comprehensions vs loops08/12/2025
  13. Expert: designing reusable components across domains09/12/2025
  14. Introduction to unit testing in Python09/12/2025
  15. Empirical: recursion vs iteration and memory impact10/12/2025
  16. Best practices for imports and packaging in monorepos11/12/2025
  17. Tools: Snyk, Dependabot, and Checkov11/12/2025
  18. Expert: model capacity and overfitting trade-offs12/12/2025
  19. Tools: abc, dataclasses, strategy helpers12/12/2025
  20. Introduction to GRASP design principles13/12/2025
  21. Best practices: clean commit history and branching models14/12/2025
  22. Tools: aiohttp and anyio for async workflows15/12/2025
  23. Expert: designing ethical governance frameworks for AI15/12/2025
  24. Best practices for consistent style with PEP816/12/2025
  25. Empirical: benchmarks of Cython, Numba, and PyPy17/12/2025
  26. Expert: interactive pipelines and parametrized runs18/12/2025
  27. Best practices: scalable architectures for data systems19/12/2025
  28. Introduction to SOLID principles in Python19/12/2025
  29. Empirical: pyinstaller vs nuitka vs cx_Freeze20/12/2025
  30. Introduction to developer productivity fundamentals23/12/2025
  31. Tools: functools, itertools, toolz24/12/2025
  32. Best practices for designing generative tests24/12/2025
  33. Expert: chaos engineering for resilient ML infrastructure25/12/2025
  34. Empirical: relative vs absolute imports in large codebases25/12/2025
  35. Expert: Bayesian regularization and priors26/12/2025
  36. Best practices: reactive and event-driven observer systems26/12/2025
  37. Tools: abstract base classes and dataclasses for GRASP27/12/2025
  38. Introduction to fairness in machine learning27/12/2025
  39. Expert: idioms for clean API and operator overloading28/12/2025