Engineering Remote Full-time Senior

Detection Engineer – Fraud Scoring

VerifyStack $190K – $260K + equity Remote

About the Role

Join our Engineering team to design, build, and optimize the deterministic scoring engine that powers VerifyStack's fraud detection. You'll work on signal processing, behavioral biometrics, anomaly detection, and rule-based scoring — engineering systems that run at the edge with sub-50ms latency while adapting to evolving attack vectors.

What You\'ll Do

Design and implement deterministic scoring algorithms for real-time fraud detection — signal weighting, threshold calibration, and anomaly scoring
Build and maintain the signal processing pipeline: feature extraction, normalization, and multi-layer fusion
Implement the feedback loop system that ingests ground-truth labels and continuously recalibrates scoring weights
Develop behavioral biometric analysis modules (typing dynamics, mouse/touch patterns, device interaction signals)
Research adversarial techniques — understand how attackers use generative AI and build defenses against evasion
Optimize detection logic for edge deployment: efficient algorithms and WebAssembly modules for sub-50ms scoring
Collaborate with backend engineers on signal stores, scoring infrastructure, and observability
Publish findings internally and externally (blog posts, conference talks, whitepapers)

Requirements

M.S. or Ph.D. in Computer Science, Statistics, or a related field (or equivalent industry experience)
4+ years of experience building production detection / scoring systems, ideally in fraud/risk/security
Strong proficiency in TypeScript and systems-level thinking — our entire stack is TypeScript from edge to database
Experience with real-time signal processing and low-latency scoring at scale
Deep understanding of classification metrics (precision, recall, AUC-ROC, calibration) and class-imbalanced datasets
Familiarity with feature engineering for sequential / behavioral data
Ability to communicate complex technical concepts clearly to non-technical stakeholders

Nice to Have

Experience with adversarial attack patterns, red-teaming, or security research
Knowledge of device fingerprinting, browser APIs, or network signal analysis
Published research in fraud detection, anomaly detection, or security engineering
Experience with WebAssembly, edge computing, or browser-based signal collection
Familiarity with consortium / federated threat intelligence approaches

Tech Stack

TypeScriptNext.jsNode.jsPostgreSQLRedis/KVPrismaVercel EdgeWebAssembly

Benefits & Perks

Competitive salary ($190K–$260K) plus meaningful equity
Fully remote — work from anywhere
Unlimited PTO with a 3-week minimum encouraged
$2,000 home office stipend
$1,500 annual learning & conference budget
Comprehensive health, dental, and vision insurance
Mental health support
401(k) matching

Interested?

Send your resume and a brief note about why you\'re excited about this role. We review every application.

Apply via Email

Or reach out on LinkedIn

About the Role

What You\'ll Do

Design and implement deterministic scoring algorithms for real-time fraud detection — signal weighting, threshold calibration, and anomaly scoring

Build and maintain the signal processing pipeline: feature extraction, normalization, and multi-layer fusion

Implement the feedback loop system that ingests ground-truth labels and continuously recalibrates scoring weights

Develop behavioral biometric analysis modules (typing dynamics, mouse/touch patterns, device interaction signals)

Research adversarial techniques — understand how attackers use generative AI and build defenses against evasion

Optimize detection logic for edge deployment: efficient algorithms and WebAssembly modules for sub-50ms scoring

Collaborate with backend engineers on signal stores, scoring infrastructure, and observability

Publish findings internally and externally (blog posts, conference talks, whitepapers)

Requirements

M.S. or Ph.D. in Computer Science, Statistics, or a related field (or equivalent industry experience)

4+ years of experience building production detection / scoring systems, ideally in fraud/risk/security

Strong proficiency in TypeScript and systems-level thinking — our entire stack is TypeScript from edge to database

Experience with real-time signal processing and low-latency scoring at scale

Deep understanding of classification metrics (precision, recall, AUC-ROC, calibration) and class-imbalanced datasets

Familiarity with feature engineering for sequential / behavioral data

Ability to communicate complex technical concepts clearly to non-technical stakeholders

Nice to Have

Experience with adversarial attack patterns, red-teaming, or security research

Knowledge of device fingerprinting, browser APIs, or network signal analysis

Published research in fraud detection, anomaly detection, or security engineering

Experience with WebAssembly, edge computing, or browser-based signal collection

Familiarity with consortium / federated threat intelligence approaches