SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 180,000 - 300,000 / annual
Traversal is an AI Site Reliability Engineer platform trusted by major enterprises to troubleshoot, remediate, and prevent complex production incidents. The company is building an AI agent lab for the enterprise, with a team including researchers from MIT, Harvard, and Berkeley alongside engineers from top firms like Citadel Securities, Datadog, and Perplexity.
As an AI Research Engineer, you'll build the infrastructure systems that enable the research team to develop and test autonomous agents that diagnose and resolve production incidents at scale. Your core responsibility is eliminating bottlenecks in the research process—not ideas, but the infrastructure to run experiments, validate results, and scale prototypes.
Key responsibilities include: designing and building experiment harnesses and evaluation pipelines that let researchers test agents against real customer incident data; creating tooling that makes agent behavior transparent and debuggable; developing systems to take promising prototypes from single runs to production-scale deployments; collaborating with researchers on experiment design and methodology; and prototyping improvements to prompting strategies, reasoning workflows, and tool-use policies for agents working with observability data.
You'll work at the intersection of research and production engineering. This is a backend engineering role for someone with strong experimentation instincts—ideally someone who has run experiments as side projects and wants that to be their primary focus. You'll partner closely with AI engineers, infrastructure teams, and product leads to translate research into customer-facing capabilities.
Required: 5+ years of software engineering with deep backend and scaled systems experience; hands-on experimentation background; proven ability to thrive in fast-paced startups with shifting priorities; strong Python and FastAPI skills; AWS deployment experience (ECS/Kubernetes); Postgres and S3 proficiency; observability and monitoring expertise.
Nice-to-have: research engineer background or prior embedded research team experience; agentic AI systems building; large-scale data-driven applications; AI/LLM product familiarity; evaluation and benchmarking tooling experience.
The role is fully in-office in New York, 5 days per week, near Madison Square Park. Traversal offers competitive salary, equity, health insurance, flexible time off, and a collaborative, high-impact team environment.
