Principal Engineer - Supply Chain Planning

THE ROLE

Principal Engineer - Supply Chain Planning

Quince is building its own supply chain planning platform from scratch because the model we operate doesn’t fit anything off the shelf. Our supply chain runs factory-direct, at high frequency, with short lead times, across a growing number of vendors, fulfillment centers, and markets worldwide. The planning and forecasting layer that coordinates all of this is being rebuilt, and as Principal Engineer for Supply Chain Planning Tools, you will build it.

This role is about building the production systems that make data science real, specifically the platform infrastructure that takes forecasting models and optimization algorithms from development into the weekly cadence that runs the business. You’ll need to understand the science well enough to partner with the people who do it, translate it into reliable systems, and give operators the controls to work alongside it and override it when needed.

This is a 0-to-1 role. You won’t inherit a system, but a set of business workflows. You’ll design and build one, with a small and highly capable team, starting with demand forecasting and planning infrastructure and expanding into logistics and warehouse optimization. 

The platform we'll build is AI-native by design. AI-augmented operator interfaces, LLM-aided observability, agentic workflows over planning data, and AI-assisted incident triage are first-class capabilities on the roadmap, not afterthoughts. The Principal Engineer will set the bar for both — how the team builds with AI, and how AI shows up in what we ship.
The ideal candidate is a seasoned production engineer who has spent meaningful time in the orbit of data science and ML, not as a practitioner, but as the person who makes it work in production. They have built model pipelines, experimentation frameworks, feature infrastructure, and operator tooling that bring algorithmic systems to life at scale. They can walk into a room with supply chain practitioners, understand their problems at a strategic level, and independently determine what to build. They don’t wait for a PM to translate, and they don’t wait for a large team to start executing.

They thrive in environments where strategy, innovation, and decision-making are intentionally distributed, where candor, speed, and data are highly valued, and colleagues at all levels hold each other to unusually high standards on behalf of Quince customers.

Responsibilities:

• AI native Platform Architecture & Build
  • Architect and build Quince’s proprietary supply chain planning platform from the ground up to be multi-vendor, multi-modal, multi-market, and built to scale.
  • Design and own the full model pipeline lifecycle, including feature engineering, forecasting tournament framework, evaluation, deployment, monitoring, and refresh, and the experimentation framework that lets scientists iterate safely in production.
  • Build integrations with vendor management, order management, and inventory platforms so the planning system sits at the center of the weekly ordering cadence
  • Set the standard for how the team builds with AI — coding assistants, AI-generated tests, AI-augmented data exploration — and hold the line on quality of AI-generated output through review rigor and refactoring. 
  • Architect AI-augmented capabilities directly into the platform: LLM-aided observability and root-cause analysis, natural-language operator interfaces over planning data, agentic workflows for routine planning tasks. 
  • Partner with the science team on the infrastructure that AI-driven models need, such as vector stores, prompt versioning, and evaluation harnesses for LLM-based components, so AI-driven science can run reliably in production alongside statistical and ML approaches.
• Operator Tooling
  • Build the operator-facing layer that makes the platform usable: dashboards, override workflows, audit trails, and alerts that translate model outputs into decisions a planner can act on
  • Design observability systems that surface model drift, data quality issues, and forecast failures before they propagate into bad orders or stock-outs
  • Build the feedback loops that let operator overrides inform and continuously improve future model performance
    
• Business Stakeholder Partnership
  • Work directly with Quince’s planning team and business leadership as a thought partner, synthesizing their operational expertise with your engineering judgment to determine what to build and in what sequence.
  • Translate business problems, such as in-stock gaps, demand volatility, vendor reliability, and fulfillment split rates, into precise engineering specifications.
  • Educate stakeholders on system capabilities and trade-offs, building the trust that lets automation and human judgment work in genuine partnership
• Technical Leadership
  • Set the technical direction for the planning tools domain, making architectural decisions that scale with Quince’s trajectory across geographies and order-of-magnitude growth
  • Partner with engineering teams across logistics, warehouse, and supply chain infrastructure to ensure the planning platform integrates cleanly into the broader ecosystem
  • Provide technical direction to a globally distributed engineering team, establishing architecture and standards that engineers across geographies can execute within
  • Recruit and grow a small, elite team around you as the charter expands

Required: 

• • 12+ years of software engineering experience with significant time building data-intensive, ML-adjacent, or science platform systems in production.
  • Demonstrated 0-to-1 platform ownership, taking a greenfield charter from architecture through production at the pace a fast-growing business demands.
  • Strong production ML and AI platform engineering experience: feature pipelines, model serving, experiment frameworks, monitoring, drift detection, and the infrastructure modern AI-driven systems need (vector stores, prompt and trace versioning, LLM evaluation harnesses). Not just architect — be able to roll up your sleeves and build.
  • AI-native engineering practice. You can speak specifically to where AI tooling has changed how you ship, such as code generation, test authoring, and AI-augmented platform features you've put into production, and where you held quality standards against AI assistance because review rigor was needed.
  • Science literacy as a collaborator; you understand forecasting and optimization models well enough to build great infrastructure around them and partner with scientists as a technical peer.
  • Strong software engineering fundamentals across the full stack: data systems, API design, pipeline orchestration, and production operations
  • Strong communication skills, able to work closely with business and product stakeholders to understand requirements, and translate it into an architectural blueprint and roadmap

Preferred:

• Experience in supply chain planning, demand forecasting systems, inventory optimization, or logistics operations is strongly preferred; candidates from adjacent domains who can rapidly develop supply chain fluency will also be considered
• Experience providing technical direction to geographically distributed engineering teams; hands-on ML or data science background is a genuine plus