Staff Software Engineer – Fullstack (AI/ML)

Location: Hybrid, 3 days per week in the Dublin office, Dublin, California, United States

Salary: $208K–$299K per year

Company

Snowflake provides a data platform and is developing AI-native applications and agents for enterprise workflows.

What you will do

• Lead the design, delivery, and scaling of full-stack applications and agentic systems for the finance organization.
• Architect, build, and deploy AI agents that plan, call tools, and execute workflows for contract review, reconciliation triage, exception resolution, and close-task automation.
• Own agent harnesses covering control loops, tool definitions, context and memory management, permissions, sandboxed execution, retries, timeouts, and error recovery.
• Implement guardrails, human-in-the-loop approvals, evaluations, observability, tracing, logging, cost, and latency monitoring for agents operating on financial data.
• Partner with Revenue Accounting on ASC 606 revenue recognition, contract processing, revenue schedules, contract modifications, and month-end close.
• Build ERP, billing, CRM, reconciliation, controls, and exception-handling integrations with SAP and Oracle systems; set engineering standards and mentor engineers.

Requirements

• Bachelor’s or Master’s degree in any field.
• 8+ years of software development experience.
• Production experience shipping agentic or LLM-powered applications.
• Hands-on experience with SAP or Oracle financial modules, ideally in revenue accounting or order-to-cash.
• Working knowledge of revenue recognition, including ASC 606 or IFRS 15, contract modifications, and month-end close.
• Strong SQL and database fundamentals, modern software development practices, problem-solving, analytical skills, and communication skills.
• Ability to partner directly with Accounting and Finance.
• Experience designing agents with tool or function calling, multi-step planning, orchestration, retrieval-augmented generation, semantic search, and grounding on structured and unstructured enterprise data.
• Experience building evaluation sets, measuring agent accuracy, implementing role-based access, auditability, approval gates, failure handling, regression suites, golden datasets, LLM-as-judge workflows, and trace replay.
• Experience with context windows, summarization, compaction, memory, data and instruction selection, safe tool and MCP server design, permission models, sandboxed execution, and approval checkpoints.

Nice to have

• Familiarity with Snowflake Cortex Agents, Model Context Protocol, LangGraph, or similar agent frameworks and platforms.
• Experience with AI models for code generation, including Claude or GPT-Codex, structured prompts, reusable workflows, AI model skills and profiles, Claude co-work, prompt engineering, reasoning workflows, and human-in-the-loop systems.

Culture & Benefits

• AI-native, experimental environment focused on rapidly testing emerging capabilities.
• Collaborative work at the intersection of engineering, finance, and agentic AI.
• Focus on creativity, technical excellence, continuous learning, and professional growth.
• Hybrid work with 3 days per week in the Dublin office.