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.
Sequential peeking: why stopping at significance misleads
Repeatedly checking a fixed-horizon p-value and stopping as soon as it falls below 0.05 changes the testing procedure. The ordinary 5% false-positive interpretation does not automatically survive that stopping rule.
Serve a model with an explicit prediction schema
A model service needs a contract for inputs and outputs. Column names alone do not define acceptable ranges, category values, required identifiers or behavior for unexpected fields. Validate before preprocessing so malfo
Sessionize event data with a clear inactivity rule
Sessionization groups a user's events into visits or activity periods. A common rule starts a new session when the gap from the previous event exceeds a chosen inactivity threshold. The threshold, identity key and exact
Set a handover boundary between delivery and support
Handover does not end responsibility instantly, but an undefined support promise traps delivery engineers and confuses the receiving team. The boundary belongs in scope before launch.
Set a rollback threshold from user-visible failures
CPU and error-rate alarms do not answer whether an AI release is harming users. Rollback criteria should use task outcomes and treat some events, such as unauthorized disclosure, as immediate triggers.
Set per-user quotas for expensive AI operations
One “request” might mean a short classification or a research workflow with retrieval, multiple model calls and tools. Counting both equally makes a quota easy to bypass and hard to budget.
Set retry and timeout budgets across agent steps
Three steps with three retries and long timeouts can exceed a user’s deadline even when every component follows its own default. Start from the end-to-end budget and allocate within it.
Shadow deployment before switching model traffic
Shadow deployment sends a copy of production-like requests to a candidate while the champion continues to control responses. It can reveal schema, latency and prediction differences without changing user-facing decisions
SHAP values: explain a prediction without claiming causation
SHAP methods use Shapley-value ideas to allocate a model prediction among features relative to a baseline. They explain the specified model under a specified background and feature-handling assumption. They do not show w
Show uncertainty in a management chart
Show uncertainty in a management chart when it can change the decision. Label the type of uncertainty and the population it concerns. A range is useful only if the reader knows whether it represents sampling uncertainty,
Silhouette scores: useful signal and common misinterpretations
The silhouette coefficient compares how tightly a row sits in its assigned cluster with how close it is to the nearest other cluster. Values range from -1 to 1. The average is useful, but it can hide small, unstable or p
Simpson's paradox in channel conversion rates
An overall conversion ranking can reverse the ranking within every segment when channels contain different segment mixes. Inspect counts and denominators before concluding that one channel performs better for comparable
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.