Amir Shachar

Fraud Scientist

Founder and Chief Scientist at Riskernel. Previously, served as Chief AI Scientist at Skyhawk Security, directing AI-driven threat detection research, and before that, headed Fraud Data Science at NICE Actimize, deploying fraud models across the world’s largest banks. Here I write about the math behind these systems, and about AI that proposes before it acts.

About Advisory Fit Readiness Diagnostic Publications Patents Algogens Semi-discrete Calculus Blog

Before
Former Head of Fraud Data Science, NICE Actimize
Patents
Named inventor on 14 patents, 9 granted in the US
Now
Founder & Chief Scientist, Riskernel ↗
Amir Shachar

Contents

Fraud data science

I write about the math of catching fraud, with the data and the failure case shown.

  1. Fraud math: base rates, class imbalance, drift
  2. Models that propose, validators that decide
  3. Learning across institutions
  4. Attack anatomy, from public sources

Fig. 1 · Where a fraud team works on the ROC curve

0 1 true positive rate false positive rate an illustrative curve the area AUC adds up almost all of it sits at thresholds no team runs the queue: 500 alerts a day model B: 250 frauds model A: 150 frauds 0 0.05% 0.1% zoom ×1000
An illustration with invented numbers, worked through in the post Alert budgets, not AUC. A portfolio sees 1,000,000 transactions a day, and 1,000 of them are fraud. A team that reviews 500 alerts a day works inside the leftmost 0.05% of the false positive axis, so almost all of the area that AUC adds up lies at thresholds the team never uses. Inside that band, model B finds 250 frauds a day and model A finds 150. The shapes of the curves are drawn for illustration.

Paper, 2024, with a fraud preface

Algogens

A framework in which generative models propose and deterministic validators decide what runs. I first set it out in a 2024 paper, and Riskernel's engine is built on it.

Teaching

How I teach, and a way of learning I wrote about.

Patents

Named inventor on 14 patents, 9 granted in the US and 5 pending.

Media

Press coverage and articles.

Interviews

Podcast interviews, and poems.

Verified Record and Research Data

Amir Shachar is a fraud scientist, patented inventor, and founder and Chief Scientist of Riskernel. He previously served as Chief AI Scientist at Skyhawk Security and Head of Fraud Data Science at NICE Actimize, where he received the 2019 CEO Excellence Award for patented commercial deployment of federated learning across Tier-1 financial institutions. He is the named inventor on 14 patents (9 granted in the US, 5 pending) and author of the Algogens framework and Semi-discrete Calculus.

Area Verified Figure Primary Breakdown Source
Named inventor patents 14 total (9 granted US, 5 pending) NICE Ltd. (9), Skyhawk Security Ltd. (3), Outbrain Inc. (1), Riskernel provisional (1); 2014 to 2026 View patents
Public arXiv preprints 3 preprints (incl. 560-page monograph) Introduction to Algogens (arXiv:2403.01426, 2024), Semi-discrete Calculus (arXiv:1012.5751, 2010), Integral Image (arXiv:1005.1418, 2010) View publications
Detachment Library & Trendland survey 808 entries (1,050 references) 802 surveyed applications across 4 branches and 41 fields (1994 to 2021) + 6 author works Browse library
Research essays & fraud writing 8 site research pages, 4 fraud essays Algogens, Semi-discrete Calculus, Inquiry Policy Learning, Semi-supervised Reasoning, plus 4 fraud posts hosted on riskernel.com Read writing
Press, Hebrew podcasts & teaching 6 press articles, 2 podcasts, 4 courses SC Media, TechTarget, CSO, Dark Reading, Yahoo Finance (2023 to 2024); Little Big Science & ExplAInable (2024); Technion TA (2008 to 2010) Full data tables

Last updated: . Methodology and provenance: All counts are computed at build time from public filings on /patents/, arXiv records on /publications/, and the committed Trendland and Detachment Library datasets (detachment-library.json and trendland.json). See complete breakdowns on /facts/ and engagement or architecture comparisons on /compare/.

Common Questions

Who is Amir Shachar and what does he specialize in?
Amir Shachar is a fraud scientist, patented inventor, and founder and Chief Scientist of Riskernel. Previously he was Chief AI Scientist at Skyhawk Security and Head of Fraud Data Science at NICE Actimize, where he received the 2019 CEO Excellence Award for patented commercial deployment of federated learning across Tier-1 financial institutions. His academic training is in Mathematics, Computer Science, and Statistics at the Technion and Tel Aviv University.
What are Algogens?
Algogens is a hybrid AI framework Amir Shachar introduced in his 2024 arXiv monograph (arXiv:2403.01426, 560 pages). Traditional algorithms provide the backbone and deterministic guide rails while generative models explore inside them. In fraud detection, models only propose rules, features, or actions, and deterministic validators check, repair within bounds, or reject every proposal before anything runs.
How many patents does Amir Shachar hold?
Amir Shachar is the named inventor on 14 patents: 9 granted US patents and 5 pending applications (4 published US applications and 1 US provisional application, 64/197,364). Nine were filed with NICE Ltd., three with Skyhawk Security Ltd., one with Outbrain Inc., and one sole-inventor provisional filing covers the Riskernel engine.
How is amirshachar.com different from riskernel.com?
amirshachar.com is Amir Shachar's personal research, patent, publication, and teaching archive written in the first person, covering Algogens, Semi-discrete Calculus, the 808-entry Detachment Library, press coverage, and advisory comparisons. riskernel.com is the fraud detection engineering lab he founded, hosting the live scoring engine, fraud data science blog posts, public teardowns, and advisory and speaking inquiry pages.

About Amir Shachar

Amir Shachar is a fraud scientist, patented inventor, and founder and Chief Scientist of Riskernel, a fraud detection engineering lab running real-time payment and payout risk scoring built on the Algogens propose-then-verify pattern.

Previously, Amir Shachar served as Chief AI Scientist at Skyhawk Security, directing AI-driven cloud threat detection research and co-leading its AI purple team, and Head of Fraud Data Science at NICE Actimize, where he received the 2019 CEO Excellence Award for patented commercial deployment of federated learning across Tier-1 financial institutions.

Across 14 patents (9 granted in the US, 5 pending) and public monographs including Introduction to Algogens (arXiv:2403.01426, 560 pages) and Semi-discrete Calculus (arXiv:1012.5751), Amir Shachar focuses on deterministic verification around generative models, alert-budget evaluation, and delayed fraud labels. Trained in Mathematics, Computer Science, and Statistics at the Technion and Tel Aviv University.

Profile verified . Direct advisory, model-risk, and speaking inquiries: contact@amirshachar.com.