Senior/Staff Applied AI Researcher, Growth

Quince•Palo Alto, CA

About The Position

We are seeking a Senior/Staff Applied AI Researcher, Growth to serve as an architect of Quince’s end-to-end agentic infrastructure for marketing execution and user growth. In this high impact individual contributor role, you will transition the company away from manual optimization loops by designing a software layer that autonomously handles creative asset generation, event-driven multi-agent workflows, and cross-platform campaign management and optimization. With 0-to-1 autonomy to deploy the frontier of LLMs and multi-agent frameworks into production financial and operational workflows, your work will directly impact the top and bottom lines of Quince as we continue to rapidly scale efficiently.

Requirements

8+ years of technical experience, including at least 1–2 years of corporate or industry experience.
Expert-level command of Python and SQL, with demonstrated mastery of advanced LLM integration frameworks and data serialization patterns.
Experience operating independently to untangle loose, ambiguously defined algorithmic or statistical optimization problems without relying on pre-existing abstraction layers.
Demonstrated impact shipping real-world software and observing data-driven revenue and operational outcomes.

Nice To Haves

Experience building custom developer velocity extensions or generative asset platforms within direct-to-consumer (DTC), retail, or e-commerce environments.
Deep familiarity with advertising networks, performance marketing data structures, programmatic bidding models, or real-time event-driven architectures.
An advanced degree (Masters or PhD) in computer science, mathematics, statistics, physics or a highly quantitative field.

Responsibilities

Serve as an architect and technical owner of Quince's end-to-end multi-agent event streams and automated systems driving brand category scaling.
Collaborate with the existing team of DS, DA and ML researchers to bring the vision to life.
Write clean, highly scalable service-layer code to maximize Large Language Models (such as Claude) and advanced prompt validation tools for real-time operational execution.
Design and scale custom automated content loops and multi-modal asset pipelines that autonomously synthesize high-quality marketing copy, graphics, and video layouts.
Build and refine autonomous decisioning pipelines that integrate with data matrices to dynamically model and optimize real-time asset deployment across platform networks.
Establish rigorous confidence limits, prompt versioning protocols, and automated fallback/rollback frameworks to handle complex LLM failure modes safely under heavy web scale.
Partner closely with Product Management, Data Science, and Engineering squads to define unified catalog schemas and eliminate cross-platform ingestion and data latency bottlenecks.
Improve ads delivery efficiency by 20% in the first 6 months, while scaling volume.