Senior Software Engineer II - Applied AI and Evaluations (Remote Eligible)

Smartsheet

1d•Remote

About The Position

For over 20 years, Smartsheet has helped people and teams achieve–well, anything. From seamless work management to smart, scalable solutions, we’ve always worked with flow. We’re building tools that empower teams to automate the manual, uncover insights, and scale smarter. But more than that, we’re creating space– space to think big, take action, and unlock the kind of work that truly matters. Because when challenge meets purpose, and passion turns into progress, that’s magic at work, and it’s what we show up for everyday. Smartsheet is building the next generation of AI-powered work management through SmartAssist, our intelligent agent platform. As we scale from early demos to production-grade agents, quality is the critical frontier and we are looking for a Agent Quality Engineer to own it. This is not a QA role. It's a deeply technical, high-autonomy position at the intersection of LLM evaluation, prompt and context engineering, and retrieval-augmented generation. You will diagnose why our agents fail, design the systems that catch regressions, and drive measurable improvements across our orchestrator and subagent fleet. You will work closely with our Agent Engineering and AI Platform teams, embedded in a team that has already shipped evaluation infrastructure on Databricks/MLflow and is building toward a mature Agent Development Lifecycle (ADLC).

Requirements

8+ years of software engineering experience, with at least 2 years working directly with LLMs in production
Deep, hands-on experience with prompt engineering and context engineering, you understand how model behavior changes with framing, structure, and input design
Strong working knowledge of RAG architectures: chunking strategies, embedding models, retrieval evaluation, and failure diagnosis
Experience building or extending LLM evaluation frameworks, you have designed scorers, worked with golden datasets, and thought carefully about what good looks like
Fluency in agent system design, you don't need to own the architecture, but you can engage as a peer on architectural tradeoffs that affect quality
Strong Python skills; comfortable working in data-heavy environments (Databricks, Delta tables, or equivalent)
Ability to communicate complex quality findings (written and verbal) to both technical and non-technical stakeholders, you can explain what’s broke, why it matters, and what needs to happen next without losing the room
Strong cross-functional judgment, you know when to escalate, when to resolve independently, and how to build credibility across engineering, product, and AI platform teams
A bias for clarity in ambiguous situations, when failure modes are murky and trade-offs are real, you bring structure and a clear point of view rather than waiting for consensus
Legally eligible to work in the U.S. on an ongoing basis
BS or MS in Computer Science, a related field, or equivalent industry experience

Nice To Haves

Experience with MLflow or similar experiment tracking platforms
Familiarity with CI-integrated evaluation pipelines
Experience with multi-agent orchestration frameworks
Prior work in an Applied AI or LLMOps function within a product company

Responsibilities

Own agent quality end-to-end: diagnosis, improvement, and validation across SmartAssist's orchestrator and subagents
Identify failure modes across quality dimensions factual accuracy, completeness, tone, actionability, and latency and prioritize what to fix
Drive quality improvements through prompt engineering, context engineering, and RAG retrieval tuning
Extend and mature our evaluation framework: scorers, golden datasets, regression gates, and online evaluation for production traffic
Close the feedback loop ensure that every change has a measurable, attributable quality signal
Collaborate with our Agent Architecture lead to distinguish quality problems that require prompt/context solutions from those that require structural fixes
Establish repeatable methodology that scales beyond any single agent or subagent

Benefits

Employer subsidized medical/vision and dental coverage for full-time employees
401k Match to help you save for your future (50% of your contribution up to the first 6% of your eligible pay)
Monthly stipend to support your work and productivity
Flexible Time Away Program, plus Sick Time Off
US employees are automatically covered under Smartsheet-sponsored life insurance, short-term, and long-term disability plans
US employees receive 12 paid holidays per year
Up to 24 weeks of Parental Leave
Personal paid Volunteer Day to support our community
Opportunities for professional growth and development including access to Udemy online courses
Company Funded Perks, including a counseling membership, local retail discounts, and your own personal Smartsheet account
Teleworking options from any registered location in the U.S. (role specific)