Learn the work behind Data, AI & Forward Deployed Engineering
Practical explanations, career decisions and reproducible workflows. Read the reasoning, inspect the evidence and follow the next skill into a real programme.
Airflow DAG design: separate orchestration from business logic
An orchestrator should coordinate dependencies, retries, schedules and observability. Embedding all transformation logic inside DAG definitions makes local testing and reuse harder. Put business logic in versioned functi
ALL versus ALLSELECTED in a share-of-total measure
A share-of-total measure is defined as much by its denominator as its numerator. Decide whether a category's share should be measured against every category or only the categories included in the reader's selection. ALL
Analyze appointment no-shows without making clinical claims
A no-show is an administrative outcome under a defined appointment policy. It should not be inferred merely because a service-end timestamp is missing. Cancellations, future appointments and unresolved records need separ
Analyze call-centre service levels with abandonment rules
A service-level percentage is meaningful only with its waiting-time threshold and denominator. Including short abandons, excluding them or counting only answered contacts can produce different percentages from the same q
Analyze marketplace seller performance with comparable cohorts
A seller with no recorded refunds may simply have newer orders whose refund windows are still open. Compare outcomes at compatible ages and within relevant product groups before interpreting a seller difference.
Analyze operating expenses without changing the cost taxonomy
A category rename can look like a new expense if the report compares periods under different classification rules. Before explaining a cost increase, apply a consistent taxonomy or build a clear bridge between the old an
Analyze repeat purchases when customer identity changes
A repeat-purchase metric depends on what the system considers one customer. If an anonymous shopper later creates an account, two observed identifiers may represent one person. Counting them separately can hide repeat be
Analyze search behaviour using no-result queries
A no-result search is an observed request that returns no matching items under the recorded search configuration. It can reveal missing content, vocabulary mismatch, indexing problems or restrictive filters. It does not
Analyze support contacts per active customer
Support contacts per active customer measures contact volume relative to an activity population. The share of active customers who contacted support measures incidence. They are different: one customer can create several
Anomaly detection in a seasonal metric
A seasonal metric can make ordinary peaks look anomalous and real incidents look normal. Compare Tuesday with the relevant weekly expectation before applying a threshold. One simple detector subtracts the value from seve
Anomaly detection: validate alerts before calling them fraud
Anomaly detection ranks cases that look unusual under selected features and a reference population. Fraud is a labelled behavioural and legal concept. The two can overlap, but an unsupervised score cannot establish fraud
ANOVA: what a significant result does not tell you
A significant one-way ANOVA result provides evidence against the hypothesis that all population group means are equal, under the model's assumptions. It does not identify every differing pair, establish that all groups d
Not sure which programme fits?
Tell us your background and we will map it to the right entry point — including saying so when a cheaper programme is the better fit. A counsellor replies within one working day.