Skip to main content
⚡
CodeMentor AIAll tracks
📊L3 · 6 modules · 100 lessons

Data Science Applied

pandas, polars, real datasets, dashboards

100 lessons across 6 modules: pandas + numpy for data work, cleaning & feature engineering, modeling & evaluation, A/B testing & causal inference, production ML pipelines, and deep-learning foundations (backprop, transformers, fine-tuning, MLOps registry + drift).

Browse all 100 lessons →Try the first lesson — no signup
100
lessons
6
modules
3
capstone missions
18
languages
What you'll learn

From day one

  • ✓Pandas + numpy fluency — vectorized ops, groupby, pivot
  • ✓Polars in 2026 — when to switch from pandas
  • ✓Data cleaning, missing values, outliers, encoding
  • ✓Train/test split discipline, k-fold CV, stratification
  • ✓Modeling: logistic regression → gradient boosting → calibration
  • ✓Causal inference: A/B tests, DiD, RDD basics
  • ✓MLOps: feature stores, model registry, drift detection
  • ✓Deep learning foundations: backprop, transformers, fine-tuning
  • ✓Capstone: end-to-end ML pipeline with eval gates
Curriculum

6 modules · 100 interactive lessons

#1Module 1 · Python for Data: pandas, numpy
#2Module 2 · Data Cleaning & Feature Engineering
#3Module 3 · Modeling & Evaluation
#4Module 4 · Causal Inference & A/B Testing
#5Module 5 · Production ML Pipelines
#6Module 6 · Deep Learning + MLOps
See all 100 lessons →

Frequently asked

Is this track for ML engineers or analysts?

Both, with different cuts. Modules 1-3 (pandas + cleaning + EDA) are the analyst path. Modules 4-6 (modeling + MLOps + deep learning) push into ML engineering.

Does it cover deep learning?

Yes — basics (backprop, transformers, fine-tuning) plus production MLOps (model registry, drift detection). Not a substitute for a dedicated DL course, but covers what a working data scientist actually does day-to-day.

What pandas version?

pandas 2.x with the modern arrow-backed types. We don't waste time on legacy SettingWithCopyWarning theatre.

Will I learn polars too?

Yes — comparative chapter showing where polars wins (lazy eval, larger-than-memory data). pandas is still the daily driver.

Capstone project?

End-to-end analytics: ingest messy CSV, clean, model, evaluate, ship a Streamlit dashboard. Numbers in the README come from a real eval — not made up.

First 15 lessons free, no signup

All 100 lessons unlock for $12/month or $89/year. 7-day free trial — card required, cancel in 1 click anytime in Settings.

See pricing