AI Data Scientist
The statistical and experimental reasoning that sits underneath modern AI work — distributions, hypothesis tests, causal inference, and A/B design — paired with the coding and EDA habits that turn raw data into a defensible answer, then into a model whose numbers can be explained to someone who didn't build it. Twenty-nine lessons from what the role actually is through shipping and communicating a result.
What Is a Data Scientist?What a data scientist actually does day to day, and how the role differs from "someone who's good with data."The Data Science LifecycleThe ask, data, analysis, decision, and impact loop that structures a real data science project from start to finish.The Math and Stats You Actually NeedA realistic orientation to the math and statistics foundation data science requires, and the parts you can safely skip early on.Data Scientist vs. ML Engineer vs. Data AnalystHow these three roles actually differ in scope, deliverables, and skills at a typical company, not just by title.Descriptive Statistics and DistributionsMean, median, and variance, plus how to read the shape of a distribution to know which summary actually describes your data.Probability for Data ScienceConditional probability, independence, and expectation, the specific probability tools a data scientist actually reaches for day to day.Hypothesis Testing and P-ValuesWhat a null hypothesis and a p-value actually mean, and the common misreadings that lead teams to the wrong conclusion.Confidence Intervals and SamplingHow sampling distributions and standard error let you build a confidence interval, and what that interval actually promises.Bayesian Thinking for Data ScientistsPriors, posteriors, and why framing a problem as Bayesian versus frequentist changes what your numbers actually mean.Regression for InferenceUsing linear regression to test hypotheses about relationships, not just to predict, and why standard errors matter more than the coefficient itself.Causal Inference BasicsWhy correlation isn't causation, how confounders fool regressions, and the core toolkit for recovering causal effects without a lab experiment.Designing A/B TestsHow to size, randomize, and run an A/B test that gives a trustworthy answer, and the peeking trap that invalidates most informal ones.Time Series FundamentalsTrend, seasonality, and stationarity are the three ideas that determine which models and which mistakes apply to time-ordered data.Python for Data ScienceThe Python idioms that matter for data work: records as dicts, functions over copy-pasted loops, and reading files without losing your afternoon.SQL for Data AnalysisWriting the SQL an analyst actually uses day to day: GROUP BY aggregations, joins across tables, and window functions for running totals and ranks.Pandas and NumPy WorkflowsReal dataframe workflows: loading messy data, cleaning it, merging tables, and vectorizing math with numpy instead of looping row by row.Notebooks and Reproducible AnalysisUsing Jupyter notebooks well: avoiding hidden state from out-of-order execution, structuring cells, and versioning notebooks sanely.The Exploratory Data Analysis ProcessA systematic order for looking at a new dataset before modeling — shape and types, distributions, relationships, then outliers.Data Cleaning and Missing DataWhy the right fix for a missing value depends on why it's missing, plus practical strategies for duplicates and messy data.Data Visualization PrinciplesWhat makes a chart actually communicate — choosing the right form, cutting chartjunk, and avoiding visuals that mislead.Feature Engineering BasicsTurning raw columns into model-ready features through encoding, scaling, binning, and interaction terms, without leaking the target.Supervised Learning for PredictionHow to pick a supervised learning approach for a prediction problem based on the data and the decision it supports, not just familiarity.Model Evaluation and MetricsWhy accuracy is often the wrong default metric, and how to choose precision, recall, F1, AUC, or RMSE based on what the model's mistakes actually cost.Overfitting and Cross-ValidationHow to spot overfitting in practice and use train/validation/test splits and cross-validation to get an honest read on model performance.Model Interpretability and ExplainabilityHow to explain feature importance, SHAP values, and partial dependence to a non-technical stakeholder without misrepresenting what a model actually does.AI-Assisted Data Science WorkflowsUsing AI coding assistants to speed up analysis work without letting unverified output into your conclusions.Taking Models to ProductionWhat happens to a model after the notebook, and which of those steps a data scientist actually owns.LLMs for Data ScientistsUsing large language models as an analysis tool, for SQL generation, summarization, and sense-checking results.Communicating Results and Data StorytellingStructuring a narrative around an analysis so stakeholders act on it correctly, without overclaiming what the data shows.
Foundations
- What Is a Data Scientist?What a data scientist actually does day to day, and how the role differs from "someone who's good with data."
- The Data Science LifecycleThe ask, data, analysis, decision, and impact loop that structures a real data science project from start to finish.
- The Math and Stats You Actually NeedA realistic orientation to the math and statistics foundation data science requires, and the parts you can safely skip early on.
- Data Scientist vs. ML Engineer vs. Data AnalystHow these three roles actually differ in scope, deliverables, and skills at a typical company, not just by title.
Statistics for data science
- Descriptive Statistics and DistributionsMean, median, and variance, plus how to read the shape of a distribution to know which summary actually describes your data.
- Probability for Data ScienceConditional probability, independence, and expectation, the specific probability tools a data scientist actually reaches for day to day.
- Hypothesis Testing and P-ValuesWhat a null hypothesis and a p-value actually mean, and the common misreadings that lead teams to the wrong conclusion.
- Confidence Intervals and SamplingHow sampling distributions and standard error let you build a confidence interval, and what that interval actually promises.
- Bayesian Thinking for Data ScientistsPriors, posteriors, and why framing a problem as Bayesian versus frequentist changes what your numbers actually mean.
Econometrics and experimentation
- Regression for InferenceUsing linear regression to test hypotheses about relationships, not just to predict, and why standard errors matter more than the coefficient itself.
- Causal Inference BasicsWhy correlation isn't causation, how confounders fool regressions, and the core toolkit for recovering causal effects without a lab experiment.
- Designing A/B TestsHow to size, randomize, and run an A/B test that gives a trustworthy answer, and the peeking trap that invalidates most informal ones.
- Time Series FundamentalsTrend, seasonality, and stationarity are the three ideas that determine which models and which mistakes apply to time-ordered data.
Data science coding
- Python for Data ScienceThe Python idioms that matter for data work: records as dicts, functions over copy-pasted loops, and reading files without losing your afternoon.
- SQL for Data AnalysisWriting the SQL an analyst actually uses day to day: GROUP BY aggregations, joins across tables, and window functions for running totals and ranks.
- Pandas and NumPy WorkflowsReal dataframe workflows: loading messy data, cleaning it, merging tables, and vectorizing math with numpy instead of looping row by row.
- Notebooks and Reproducible AnalysisUsing Jupyter notebooks well: avoiding hidden state from out-of-order execution, structuring cells, and versioning notebooks sanely.
Exploratory data analysis
- The Exploratory Data Analysis ProcessA systematic order for looking at a new dataset before modeling — shape and types, distributions, relationships, then outliers.
- Data Cleaning and Missing DataWhy the right fix for a missing value depends on why it's missing, plus practical strategies for duplicates and messy data.
- Data Visualization PrinciplesWhat makes a chart actually communicate — choosing the right form, cutting chartjunk, and avoiding visuals that mislead.
- Feature Engineering BasicsTurning raw columns into model-ready features through encoding, scaling, binning, and interaction terms, without leaking the target.
Applied modeling
- Supervised Learning for PredictionHow to pick a supervised learning approach for a prediction problem based on the data and the decision it supports, not just familiarity.
- Model Evaluation and MetricsWhy accuracy is often the wrong default metric, and how to choose precision, recall, F1, AUC, or RMSE based on what the model's mistakes actually cost.
- Overfitting and Cross-ValidationHow to spot overfitting in practice and use train/validation/test splits and cross-validation to get an honest read on model performance.
- Model Interpretability and ExplainabilityHow to explain feature importance, SHAP values, and partial dependence to a non-technical stakeholder without misrepresenting what a model actually does.
Shipping and communicating insights
- AI-Assisted Data Science WorkflowsUsing AI coding assistants to speed up analysis work without letting unverified output into your conclusions.
- Taking Models to ProductionWhat happens to a model after the notebook, and which of those steps a data scientist actually owns.
- LLMs for Data ScientistsUsing large language models as an analysis tool, for SQL generation, summarization, and sense-checking results.
- Communicating Results and Data StorytellingStructuring a narrative around an analysis so stakeholders act on it correctly, without overclaiming what the data shows.