"fraud decisioning" Jobs

160 open tech roles matching “fraud decisioning”, taken straight from company career pages — not reposted from other job boards. Most in demand right now: Python, AWS, AI/ML. Every listing is re-checked daily and closed roles are removed.

Showing 20 of 160 results

Affirm

Lead projects on the Fraud Decisioning team to enhance Affirm's fraud-fighting capabilities in a remote-first environment.

Affirm Remote US $204k–$290k/yr Published 2 days ago
Anthropic

Join Anthropic as a Staff Software Engineer to build systems that combat financial fraud in a collaborative environment.

Anthropic San Francisco, CA | New York City, NY | Seattle, WA $320k–$485k/yr Published 3 months ago
Flexible on stack
Affirm

Lead projects on the Fraud Decisioning team to enhance fraud-fighting capabilities at Affirm.

Affirm Remote Canada CA$181k–CA$241k/yr Published 2 days ago
Mercury
Mercury San Francisco, CA, New York, NY, Portland, OR, or Remote within Canada or United States $239k–$298.8k/yr Published 6 months ago
Stripe

Join Stripe as a Staff Product Manager to lead fraud and abuse prevention efforts with a focus on customer insights and product strategy.

Stripe San Francisco, Seattle, New York, US-Remote Published 6 days ago
Phantom

Build systems to safeguard users and products as a Staff Engineer in a fully remote team at Phantom.

Phantom Remote $200k–$250k/yr Published 2 weeks ago
Reddit

Shape Reddit's threat response frameworks as a Staff Threat Analyst focused on advertiser fraud in a fully remote role.

Reddit Remote - United States $167.5k–$234.5k/yr Published 3 weeks ago
Flexible on stack
Affirm

Lead technical strategy for backend identity decisioning systems in a remote-first environment at Affirm.

Affirm Remote US $204k–$290k/yr Published 1 week ago
Flexible on stack
Affirm

Lead technical strategy for backend identity decisioning at Affirm with a focus on international markets.

Affirm Remote Canada CA$181k–CA$241k/yr Published 1 week ago
Flexible on stack
Forter

Join Forter as a Staff Engineer to design and scale backend systems for real-time fraud decisioning and APIs.

Forter Israel - Tel Aviv Published 2 months ago
Flexible on stack
Anthropic

Join Anthropic as a Staff+ Software Engineer to build machine learning systems that detect and prevent account abuse at scale.

Anthropic San Francisco, CA | New York City, NY $320k–$485k/yr Published 1 week ago
Flexible on stack
MoonPay

Own the decisioning system for real-time transaction scoring in a high-impact fraud detection role at MoonPay.

MoonPay London - Hybrid Published 1 month ago
Flexible on stack 70% coding
Block

Build production ML systems to transform customer behavior into trusted signals for decision-making at Block.

Block Bay Area, CA, United States of America $276.8k–$415.2k/yr Published 11 months ago
Flexible on stack
Stripe

Own technical direction for screening and detection systems in a team focused on financial crimes at Stripe.

Stripe Toronto Published 2 months ago
Flexible on stack
SentiLink

Join SentiLink as a Staff Data Scientist to build fraud detection models and drive financial risk products in a fast-paced environment.

SentiLink United States $220k–$260k/yr Published 7 months ago
Flexible on stack
SentiLink

Join SentiLink as a Staff Applied ML Scientist to develop fraud detection models and drive innovative identity verification solutions.

SentiLink United States $220k–$260k/yr Published 7 months ago
Flexible on stack
Kikoff

Join Kikoff as a Staff Machine Learning Engineer to lead ML initiatives that empower financial security for millions.

Kikoff San Francisco $337.7k–$387.2k/yr Published 1 month ago
Flexible on stack
SentiLink

Join SentiLink as a Staff Software Engineer to lead the Data Platform team in building scalable fraud detection systems.

SentiLink United States $220k–$260k/yr Published 2 months ago
Flexible on stack
Affirm

Join Affirm as a Staff Software Engineer to shape the Identity backend systems for global expansion in a remote-first environment.

Affirm Remote UK £142k–£190k/yr Published 1 month ago
Flexible on stack