Senior Software Engineer – Agent Systems

Payward is the parent company behind Kraken, NinjaTrader, Breakout, xStocks, Payward Services, and CF Benchmarks. Founded in 2011, Kraken is one of the world's longest-standing crypto platforms, trusted by over 10 million individuals and institutions globally. The Agent Systems team is a 0→1 engineering group building internal AI-powered agents to increase internal speed of execution through automation, intelligent inference, and workflow orchestration.

• 5+ years of experience building and shipping production software systems
• Strong backend engineering fundamentals including APIs, distributed systems, data modeling, and reliability
• Experience building fullstack systems with comfort across backend and occasional frontend surfaces
• Hands-on experience working with AI/ML systems, LLMs, or inference pipelines in production
• Ability to prototype rapidly while maintaining a clear path to production-grade architecture
• Strong systems thinking with ability to reason about orchestration, failure modes, and observability
• Comfort operating in ambiguous, 0→1 environments with high ownership

Nice to have:

• Experience building or operating agent-based systems or autonomous workflows
• Familiarity with prompt engineering, evaluation frameworks, or model performance tuning
• Experience integrating with internal enterprise systems such as CRM, support tools, and analytics platforms
• Background in building developer platforms or internal tooling
• Experience designing guardrails, sandboxing, or safety layers for AI systems

• Build and ship AI-powered agent systems that automate internal workflows and decision paths
• Design and implement inference pipelines that reason over user interactions across internal systems
• Prototype quickly to validate agent behaviors, then productionize what proves valuable
• Architect backend services that integrate with multiple internal APIs and data sources
• Define patterns for safe, observable, and controllable agent execution
• Collaborate cross-functionally to identify high-leverage automation opportunities
• Improve system speed, reliability, and signal quality as usage scales
• Contribute to foundational design decisions for a 0→1 agent platform

Applications are accepted on an ongoing basis unless a specific deadline is stated. Applicants may redact or remove age, date of birth, or education dates from resumes. The employer considers applicants with criminal histories in compliance with the San Francisco Fair Chance Ordinance. Employment decisions are based on merit without discrimination against protected characteristics as defined by federal, state, or local laws.