SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Klaviyo is seeking a Senior Staff Engineer to lead the technical direction of their Customer Agent platform, a conversational AI system enabling brands to serve customers at scale. This is a principal-caliber individual contributor role where you'll architect, build, and operate a production agentic system end-to-end.
You'll own the platform's hardest technical challenges: orchestration, evaluation frameworks, guardrails, memory and context management, self-healing mechanisms, and failure handling for non-deterministic systems. The role combines hands-on engineering—writing and shipping code, reviewing designs, making critical technical calls—with broader platform leadership. You'll build reusable architecture and frameworks that other teams adopt, mentor through design reviews, and partner with product, design, data science, and go-to-market leaders to shape the roadmap.
The Customer Agent team is small, ambitious, and past the demo stage. You'll ship changes against live customer traffic, measure quality with real evaluation frameworks, and treat reliability as a product feature. This is a 0-to-1 product opportunity with startup speed backed by Klaviyo's scale and resources.
Required experience: You've architected, built, scaled, and supported a production agentic system end-to-end with real orchestration, evals, guardrails, and non-deterministic failure handling. You've owned systems through launch, scale, incidents, and years of operational reality. You've learned production lessons around security, abuse prevention, and keeping non-deterministic systems reliable at scale. You've set technical and product direction for an entire product area, not just components, and can point to designs you championed that shipped and became foundations for other teams. You stay hands-on: writing code, reviewing your own decisions, debugging production issues yourself. Your past architecture and frameworks were reused across teams, unblocking multiple groups simultaneously. You communicate complex technical trade-offs clearly across engineering, product, and executive audiences.
Nice-to-haves: Multi-tenant agent platform experience at production scale, 0-to-1 product or startup experience, production evaluation systems, AI observability, or human-in-the-loop workflows for LLM products.
