Senior Software Engineer, Applied AI

Job description

Job responsibilities

• Own architecture and end-to-end implementation of production agentic products and workflows from problem definition through deployment and operation.
• Design boundaries between deterministic software, workflow engines, databases, source systems, and probabilistic agent reasoning.
• Build and review backend services, APIs, full-stack applications, and operator/user-facing tools supporting agentic workflows.
• Design agent orchestration, tool contracts, workflow state, memory, context, retrieval, approvals, escalation, retries, idempotency, rollback, and failure recovery.
• Design production data architecture, including relational schemas, state persistence, caching/search, event systems, and consistency behavior.
• Integrate AI systems with enterprise tools, internal APIs, data platforms, and operational systems while preserving permissions, auditability, and source-of-authority boundaries.
• Partner with infrastructure teams on cloud architecture, containers, CI/CD, queues, observability, tracing, security, scaling, latency, and cost.
• Establish reusable engineering patterns and reference implementations; work directly with users to convert high-value workflows into reliable products.

Job requirements

• 6+ years of relevant professional software engineering experience, or equivalent demonstrated experience building and shipping production systems.
• Demonstrated hands-on experience architecting, building, shipping, and operating production-grade AI agentic applications, products, or business workflows.
• Strong backend and full-stack engineering capability across services, APIs, databases, application state, and user-facing web applications.
• Strong programming skills in Python and at least one modern application language such as TypeScript/JavaScript, or Go.
• Strong understanding of distributed systems, databases, data modeling, transactions/consistency, event-driven architectures, and production application design.
• Strong working knowledge of cloud infrastructure, containers, CI/CD, observability, security, and production operations.
• Deep practical understanding of LLM/agent systems, including tool calling, orchestration, context, retrieval, memory, workflow state, human review, guardrails, and failure recovery.
• Experience owning real production issues such as nondeterministic failures, stale context, tool errors, duplicate actions, permission failures, model regressions, latency, and cost trade-offs.