Forward Deployed Engineer, Frontier Engineering

Overview

Forward Deployed Engineer on Snowflake's Frontier Engineering team, serving as a hands-on builder and technical partner to strategic customers to deploy production-grade enterprise AI systems.

What you'll do

• Architect, build, and deploy enterprise-grade AI solutions, including sophisticated AI agents.
• Own the end-to-end lifecycle of workstreams from prototype to production, solving customers' most complex business challenges.
• Define quality metrics, evaluation frameworks, and golden datasets, and run systematic eval loops to improve agent quality and catch regressions.
• Design, iterate, and ship high-quality code and pipelines using Python and SQL.
• Own the full implementation lifecycle for solutions, including deployment, monitoring, and optimization in secure, large-scale production environments.
• Build safety guardrails, observability, and human-review workflows for AI applications.
• Partner directly with customer data science and engineering teams as a technical resource and trusted advisor.
• Collaborate with Snowflake's Product and Engineering teams, sharing field feedback to influence the AI platform.

What you'll need

• Bachelor's degree in Computer Science, Engineering, a related technical field, or equivalent practical experience.
• 3+ years of professional software engineering experience.
• Willingness to travel.
• Proven experience building applications using LLMs, especially with technologies like RAG and agentic workflows.
• Hands-on experience defining quality metrics and running evaluations for LLM or agent systems, and using evals to systematically improve quality.
• Excellent problem-solving and communication skills, with an ability to articulate complex technical concepts to diverse stakeholders.
• Comfort with ambiguity and a desire to thrive in a fast-paced, ever-changing Generative AI environment.

Nice to have

• Experience building eval sets from production traces and synthetic data, and running structured experimentation (A/B tests, ablations, offline evals) to compare prompts, models, or agent architectures.
• Familiarity with eval and observability tooling (e.g., Braintrust, LangSmith, Arize, Weave, Promptfoo) or experience building custom eval harnesses.
• Experience with failure-mode analysis on agent or RAG systems, categorizing errors (hallucination, retrieval miss, planning failure, tool misuse) and driving each down with targeted evals.
• Hands-on experience with the MLOps lifecycle, including model deployment, monitoring, and evaluation in a cloud environment (AWS, Azure, or GCP).
• Familiarity with core data science libraries and tools (e.g., pandas, numpy, Snowpark).
• Experience in a customer-facing technical role (e.g., solutions architect, sales engineer, or professional services).
• Startup experience.

Details

• Location: New Zealand, remote.
• At least 25% of time spent onsite with Snowflake's strategic customers, including travel.