STUDY / 01
TEMPLAR-Fraud lab
Independent implementation of a routed, verifier-grounded, calibrated and cost-sensitive fraud decision pipeline with synthetic demo data and a paper fidelity audit.
Work
Each case study answers the same seven questions: what decision it supports, who uses it, what data enters, what is computed, what action is suggested, what evidence supports it, and what fails or is uncertain.
9 of 9 projects.
STUDY / 01
Independent implementation of a routed, verifier-grounded, calibrated and cost-sensitive fraud decision pipeline with synthetic demo data and a paper fidelity audit.
STUDY / 02
Independent implementation of an attention encoder with a fixed-weight extreme learning machine head and grey wolf hyperparameter search for IIoT intrusion detection, with a fidelity audit.
DEMO / 03
Deterministic discrete-event simulator of a fictional supplier-to-payment B2B workflow with idempotent events, review queues and policy comparison.
DEMO / 04
Synthetic receivables lab: three-way matching, duplicate invoice checks, payment-delay baselines and a 13-week cash forecast with intervals.
DEMO / 05
Drift, calibration and data-quality monitoring for prediction streams with a documented alert state machine and delayed-label handling.
DEMO / 06
Demand forecasting baselines with rolling-origin backtests and a synthetic replenishment simulator for service level and holding cost trade-offs.
DEMO / 07
Leakage-controlled churn-risk demonstration on synthetic B2B accounts: point-in-time features, cohort retention, calibration and capacity-limited ranking.
DEMO / 08
Complaint categorisation and review routing on the CFPB schema with a TF-IDF baseline, abstention thresholds and chronology-aware evaluation.
DEMO / 09
Cached SEC EDGAR company-facts ingestion with source-cited ratio analysis and review flags, never a fraud verdict.