SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 176,000 - 264,000 / annual
Klaviyo is seeking a Staff Engineer to set technical and product direction for a core area of their Customer Agent platform—a conversational AI system that enables brands to serve customers at scale. This is a hands-on, high-ownership role where you'll work on a 0-to-1 agentic product with startup speed backed by enterprise resources.
You will own complex, ambiguous problems across the full stack: orchestration, evaluation frameworks, guardrails, memory and context management, and failure handling for non-deterministic systems. You'll write and ship code, review designs, make architectural decisions when there's no obvious right answer, and define patterns that other engineers adopt. You'll mentor through design reviews rather than direct management, and partner with product, design, data science, and go-to-market teams to shape what gets built. You'll also evaluate emerging AI technologies and bring the ones worth betting on into production.
The team is based in Boston and works in-office. They ship changes against live customer traffic, measure quality with real evaluation frameworks, and treat reliability as a product feature. You'll be part of a small team where one engineer's work visibly shapes platform evolution.
REQUIREMENTS:
- Proven track record building, scaling, and supporting production agentic systems (not prototypes): real orchestration, evals, guardrails, and non-deterministic failure handling. You must have owned your part through launch, incidents, and post-demo operational reality.
- Deep operational experience with production agents: securing them against misuse and abuse, and the operational knowledge that comes from keeping non-deterministic systems reliable as usage grows.
- Prior ownership of technical and product direction for a significant area (not just your own component). You can point to something you designed and championed that shipped and that other engineers built on.
- Hands-on engineering: writing code, reviewing architecture decisions, and debugging production issues yourself—not delegating the hardest problems.
- Your patterns and designs are reused by other engineers; you've unblocked many others' projects, not just your own.
- Strong communication of complex technical trade-offs across engineering and product audiences; ability to turn undefined problems into plans others rally behind.
NICE TO HAVES:
- Experience with multi-tenant agent platforms at production scale where external customers configure agent behavior on shared infrastructure.
- Background building 0-to-1 products or working in startup environments.
- Track record with production evaluation systems, AI observability, or human-in-the-loop workflows for LLM-powered products.
