SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenRouter is the leading AI routing and infrastructure layer that enterprises use to access, manage, and optimize large language models across providers without lock-in or unnecessary cost. The company powers advanced AI teams by providing flexibility to move fast, scale confidently, and stay future-proof as models evolve.
You will join the Trust & Safety team as a hands-on Software Engineer responsible for building detection, enforcement, and review systems that prevent abuse, fraud, malicious users, and illegal content on the platform. Millions of LLM requests and significant payment volume flow through OpenRouter daily, making platform safety and integrity a core priority.
Key responsibilities include:
- Building and operating technical systems across signup, payments, and usage to detect abuse and fraud early
- Owning the technical enforcement pipeline end-to-end, from detection through human review to enforcement across systems
- Building internal tools for investigation and enforcement decision-making, including case queues, evidence summaries, bulk review capabilities, and alerting mechanisms
- Shipping technical solutions for content safety on the inference path, including illegal content detection and KYC/external intelligence systems
- Making abuse uneconomical through systems, tools, and heuristics for quick detection and scalable enforcement
- Investigating incidents directly in data and building analytics to establish what happened, size patterns, and distinguish real abuse from false positives
- Building defensible monitoring for risk, abuse spikes, and fraud while reducing false positives
- Setting technical direction for abuse prevention as the platform grows and defining patterns other engineers build against
- Collaborating with data scientists on feature exploration and ML model training for risk and abuse detection
This is a broad-surface-area IC role spanning the inference path, payments path, internal review tooling, and the data platform. You'll work in a small, fast-moving environment with little process between you and shipping, where your work has direct consequences for real users.
REQUIREMENTS:
- 4+ years building and operating production systems, ideally including trust & safety, fraud, payments risk, security, or anti-abuse work
- 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
- Sound judgment when working with incomplete evidence
- High agency and bias toward action; spotting problems, iterating, and shipping without waiting for tickets
- AI-forward workflow: using coding agents frequently, automating work, and holding informed opinions on tools
- Comfort in small, fast-moving environments with fluid team boundaries
- Discretion and resilience for reviewing sensitive content and handling sensitive user data
- Strong written and verbal communication to explain findings, detection rules, and systems to diverse audiences
- Motivation by adversarial problems where counterparties adapt and work continuously evolves
NICE TO HAVES:
- Experience with payments fraud tooling (Stripe Radar, chargeback/dispute flows, crypto payment risk) or identity/KYC systems
- Experience with LLM-specific abuse: jailbreaks, prompt injection, key theft/resale, shared/scraped credentials, account farming
- Experience operating large-scale analytical datastores (ClickHouse, BigQuery) and observability platforms
- Familiarity with reporting and compliance obligations for hosted AI, including illegal content reporting and model provider policies
- Active use of OpenRouter or side projects in AI products/infrastructure or developer tooling
