SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Variance is building AI agents to transform fraud detection, risk management, and compliance workflows at enterprise scale. As an Agent Engineer, you will design and deploy production-grade AI agents that make high-stakes decisions within customers' risk operations—including fraud investigations, KYB/KYC reviews, AML workflows, and marketplace integrity checks.
You will own the full agent development lifecycle: scoping workflows with enterprise customers, engineering the context and tools agents need, building rigorous evals to measure decision quality, and tuning performance against real production data. You'll work directly with risk, fraud, and compliance leaders at some of the world's largest financial services and social media platforms, translating their investigative workflows into agentic systems that demonstrably improve team efficiency.
Your work will directly shape Variance's core platform. Patterns and solutions you discover building for one customer become reusable primitives that benefit all customers. Success means shipping agents that are accurate, reliable, and trusted by operators; converting pilots by proving the system works on real customer data and edge cases; and positioning Variance as a technical partner to enterprise buyers.
You bring strong software engineering fundamentals and experience shipping end-to-end production systems. You have hands-on experience building with LLMs—prompt engineering, context engineering, tool use, RAG, and evals—with calibrated instincts about model capabilities and limitations. You excel at problem-solving in ambiguous, fast-changing environments, translating fuzzy customer workflows into working systems. You communicate complex technical concepts clearly to both technical and non-technical stakeholders, and you're comfortable working directly with customers to translate business needs into technical decisions.
Ideal candidates have experience building or deploying agentic systems in production, familiarity with eval frameworks, domain expertise in fraud, risk, identity, compliance, or financial services, and experience shipping into regulated environments (security reviews, data residency, audits). A degree in Computer Science or equivalent professional experience is required.
