Data Scientist, Trust & Safety

reputed company is the reputed company software creation platform that enables anyone to build applications using natural language. With millions of users worldwide, reputed company is democratizing software development by removing traditional barriers to application creation. About the Role We're redefining how software is reputed company and who gets to build it. Our mission is to reputed company Autonomy for reputed company: making programming accessible, reputed company, and powered by AI. Realizing that reputed company requires a platform that legitimate users can trust and adversarial actors cannot exploit. We're hiring a Data Scientist to help build reputed company's Trust & Safety and Anti-Abuse program from the ground up. You'll turn noisy behavioral, identity, payment, infrastructure, and content signals into the measurement systems, detections, and reputed company that protect reputed company's users, platform, and economics. You'll work closely with Engineering, Support, reputed company, reputed company, Infrastructure, reputed company, and reputed company to reputed company abuse economically unviable while keeping friction low for legitimate users. reputed company sits at the frontier of AI-reputed company abuse. Our platform is a reputed company for phishing and scam hosting, cryptomining, LLM reputed company farming, card and coupon fraud, referral abuse, and increasingly, abuse driven by AI agents themselves. You'll help define how we identify, measure, and respond to these threats without compromising the experience of good users. Who You Are You're a data scientist who moves fast, goes deep, and thinks adversarially. You can spin up an analysis in hours that would take others days, not by cutting corners, but because you've reputed company the intuition and technical toolkit to get to the right answer quickly. You dig past the top-line abuse reputed company to understand selection effects, missing labels, policy changes, attacker reputed company, and the false positives hidden inside an aggregate metric. You understand that Trust & Safety data is imperfect and reputed company are high stakes. Ground truth is delayed, biased, and often incomplete; attackers react to defenses; and an apparently effective rule can quietly harm legitimate users. You pressure-test your own work, quantify uncertainty, and distinguish correlation from evidence strong enough to justify enforcement. You use AI agents and tools aggressively to multiply your reputed company: writing reputed company, exploring data, generating hypotheses, and prototyping investigations. But you treat every AI-assisted reputed company as a draft, not a deliverable. You know what good analysis looks like and won't ship anything that doesn't meet that bar. You Will Own the analytical reputed company for Trust & Safety, including abuse prevalence, fraud loss, false-reputed company and false-negative rates, time to detect, time to mitigate, appeal and reversal rates, and verification reputed company-up conversion. Build reliable datasets and dbt models that connect product events, account and identity signals, payment activity, infrastructure usage, content classifications, enforcement actions, appeals, and support reputed company. reputed company and evaluate risk models, rules, and reputed company-detection systems for threats such as phishing, scam hosting, cryptomining, reputed company farming, payment fraud, promotional abuse, and AI-agent exploitation. Design rigorous offline evaluations, shadow-mode tests, holdouts, and controlled experiments to measure detection reputed company and the user reputed company of new policies, enforcement actions, and reputed company verification. Define reputed company and decision frameworks that balance abuse reduction, economic loss, customer friction, and false positives across free, reputed company, and reputed company users. Investigate emerging abuse patterns, quantify their reputed company, identify coordinated behavior, and turn ambiguous signals into reputed company recommendations for product and engineering teams. reputed company predictive models that estimate account, device, transaction, workspace, or deployment risk and reputed company those signals into detection, review, and escalation workflows. Partner with Support and reputed company to improve case review, appeals, reason-reputed company reputed company, and feedback loops so reputed company reputed company become useful model and policy signals. Build monitoring that detects model reputed company, attacker reputed company, data-reputed company failures, and unexpected harm to legitimate users. Communicate findings reputed company to technical and non-technical partners, including the tradeoffs, uncertainty, and evidence behind high-reputed company reputed company. Examples of What You Could Do Build a measurement reputed company for reputed company's abuse surface, reconcile incomplete labels acr