← Back
Home › Systems › Security and identity

fraud-detection

real-time fraud detection

Stripe Radar

SystemTechniques · 17Primitives · 55 fraud-detection Windowed stream aggregationdistributed · medium: Group events by event-time window per key, keep one partial aggregate per slice, and emit a window when the watermark passes its end.›Feature storemachine-learning · medium: One definition per feature feeds a history store for training and a latest-value store for serving, so training and inference see the same values as of the same moment.›Gradient-boosted decision trees (GB…machine-learning · medium: Add small regression trees one at a time, each fitted to the loss gradient; histogram splits over 256 bins make it the strongest tabular default.›Device fingerprintingsecurity · medium: Collect stable and volatile device signals, hash the stable core for exact lookup, fuzzy-match the rest with drift tolerance, and feed the result into a risk score.›Reputation scoringsecurity · medium: Turn per-entity good/bad event counts into a trust score that forgets old evidence, borrows strength from parent groups for new keys, and maps to allow, challenge or block.›A/B testing (online controlled expe…statistics · medium: Randomly split units into arms by a salted hash, log exposure, compare metrics with a confidence interval, and guard the result with sample-ratio, guardrail and peeking checks.›Entity resolutionstorage · medium: Decide which records describe the same real-world thing: block to cut the pair space, score pairs with a probabilistic model, cluster the matches transitively.›Union-findgraphs · low: Disjoint-set forest with union by size and path compression; merges sets and answers same-set queries in effectively constant time.›Embeddingsmachine-learning · low: A learned table mapping discrete ids to dense vectors: gather forward, sparse scatter-add backward, row-wise optimizers to keep huge tables cheap.›Count-min sketchdata-structures · low: d rows of w counters estimate any item's frequency in a stream, overestimating by at most ε·N with probability 1-δ, in fixed tiny memory.›HyperLogLogdata-structures · low: Counts distinct elements in kilobytes with about 1 % error by tracking the longest run of leading zero bits in hashed values per bucket.›Calibrationmachine-learning · low: Make predicted probabilities match observed frequencies: measure with ECE, fix post hoc with temperature, Platt or isotonic scaling on held-out data.›Evaluation metricsmachine-learning · low: Precision, recall, ROC AUC, average precision, log loss, NDCG and pass@k computed correctly, plus paired confidence intervals for comparing models.›Isolation forestmachine-learning · low: Score each record by how few random axis-aligned splits it takes to isolate it in a small forest of random trees; anomalies are few and different, so they isolate fast.›Hedged requestssystems · low: Sending a backup request after a latency percentile and taking the first answer to cut tail latencyHash aggregationstorage · low: GROUP BY via hash tables with partitioning and spill to diskHierarchical timing wheelsystems · low: Timers bucketed by expiry into wheels of slots, coarser wheels for far deadlines: O(1) insert and cancel for millions of timeouts.›Columnar file formatstorage · low: Row groups, column chunks, pages, encodings and footer stats (Parquet/ORC style) (listed in Moved to the wiki)Quantile sketchesdata-structures · low: Mergeable quantile sketches: t-digest, DDSketch, HDR histogramsLoss functionsmachine-learning · low: The scalar a model minimizes: cross-entropy for classes and tokens, MSE or Huber for values, each with a stable form and simple gradient.›Regularizationmachine-learning · low: Techniques that trade training fit for generalization: weight decay, dropout, early stopping, label smoothing and data mixing, with the exact update rules and defaults.›Fast hash functionsdata-structures · low: Non-cryptographic hashes built from 64×64→128-bit multiplies and xor-shifts; pick by key size, trust level and whether the hash is persisted.›User agent parsingsearch-sort · low: Parse user-agent strings and client hints into browser, OS and device+ 10 more techniques+ 39 more primitives

Similar systems

Share the most building blocks with this one.

ads-auction 30 %aml-monitoring 29 %model-monitoring 27 %realtime-personalization 26 %recommender 25 %

In-house systems

Named systems of this kind at large companies, with their paper or write-up.

Visa Advanced AuthorizationVisaDecision IntelligenceMastercard