Senior Software Engineer II - AI Engineering (remote Eligible)

Smartsheet Smartsheet · Seattle · United States · Engineering - Developers

Senior Software Engineer II focused on building and operationalizing production-grade LLM-powered agents and workflows within an enterprise AI platform (SmartAssist). The role involves designing agent architectures, developing RAG pipelines, implementing evaluation frameworks, and managing AI systems on cloud infrastructure. It emphasizes collaboration, mentorship, and establishing best practices for AI development and governance.

What you'd actually do

  1. Design and build production-grade LLM-powered agents and workflows within Smartsheet, including architecture decisions on system design, data pipelines, and deployment strategies
  2. Develop and optimize prompts, RAG pipelines, and agent reasoning patterns; build evaluation frameworks to measure accuracy, hallucination rates, and performance across model versions
  3. Implement and manage end-to-end AI systems on cloud infrastructure (AWS, GCP, or Azure), including monitoring, optimization, and incident response
  4. Collaborate with Engineering, Product, and cross-functional teams to translate business requirements into technical AI solutions that drive real impact
  5. Mentor and guide other engineers on applied ML best practices, modern AI development patterns, and production considerations

Skills

Required

  • 8+ years of software engineering experience
  • at least 2 years working directly with LLMs in production
  • prompt engineering
  • context engineering
  • RAG architectures
  • LLM evaluation frameworks
  • agent system design
  • Python
  • Databricks, Delta tables, or equivalent
  • communicate complex quality findings
  • cross-functional judgment

Nice to have

  • MLflow or similar experiment tracking platforms
  • CI-integrated evaluation pipelines
  • multi-agent orchestration frameworks
  • Applied AI or LLMOps function within a product company

What the JD emphasized

  • at least 2 years working directly with LLMs in production
  • Deep, hands-on experience with prompt engineering and context engineering
  • Strong working knowledge of RAG architectures
  • Experience building or extending LLM evaluation frameworks
  • Fluency in agent system design
  • Delivered measurable, validated quality improvement on at least one SmartAssist agent
  • Expanded evaluation coverage to close the most significant blind spots in our current framework
  • Established a repeatable quality improvement methodology

Other signals

  • AI agents
  • LLM-powered agents
  • Agent Development Lifecycle (ADLC)
  • production-grade LLM-powered agents
  • RAG pipelines
  • agent reasoning patterns
  • LLM evaluation frameworks
  • agent system design
  • LLMOps