Software Engineer, Trust & Safety
OpenRouter2 hours ago
Remote, United StatesMid Level
Base Salary
$215k - $285k/yr
Responsibilities
- Build and operate systems across signup, payments, and usage to detect abuse and fraud.
- Own the technical enforcement pipeline from detection and human review through restrictions across platform systems.
- Build internal investigation and enforcement tools, including case queues, evidence summaries, bulk review, enactment, and alerting.
- Develop inference-path content-safety solutions, illegal-content reporting, and KYC or external-intelligence integrations.
- Create detection heuristics, monitoring, analytics, and risk systems that enable scalable enforcement and reduce false positives.
- Investigate incidents in data and collaborate with data scientists on risk and abuse ML models.
- Set technical direction and define reusable abuse-prevention patterns for other engineers.
Requirements
- 4+ years of experience building and operating production systems, ideally in trust and safety, fraud, payments risk, security, or anti-abuse.
- Proficiency in React, TypeScript, Next.js, and JavaScript runtimes.
- Ability to write SQL against large event datasets and reason about base rates, precision, recall, and decision costs.
- Strong judgment with incomplete evidence, high agency, and comfort shipping in a small, fast-moving environment.
- Discretion and resilience when reviewing disturbing content or handling sensitive user data.
- Strong written and verbal communication skills for explaining incidents, detection rules, and systems.
- Motivation to solve evolving adversarial problems.
- Preferred experience with Stripe Radar, chargebacks and disputes, crypto payment risk, identity and KYC systems, LLM-specific abuse, large-scale analytical datastores, observability platforms, hosted-AI reporting and compliance, or AI infrastructure and developer-tooling projects.
Tech Stack
Categories
About OpenRouter
OpenRouter builds a unified API and routing layer that lets developers and enterprises access, compare, and manage large language models across multiple providers, with tooling for chat, evals, and usage controls. The company monetizes through metered API usage and enterprise plans that add governance and reliability features. Founded in 2023 and headquartered in New York, it supports hundreds of third-party models and an active developer community on Discord.