Quince

Staff Engineer - ML Infra / MLOps

Quince
Apply
2 months ago
Palo Alto, CA, USAStaff+

Base Salary

$218k - $285k/yr

Responsibilities

  • Own the end-to-end technical design of Quince’s modular and scalable ML platform, including training, serving, feature pipelines, and monitoring.
  • Build the developer experience that helps data scientists and AI researchers move from idea to production with minimal friction and high reliability.
  • Establish and implement standards for ML CI/CD, infrastructure as code, model versioning, experiment tracking, and blue-green or canary deployments.
  • Evaluate and select core platform components, including inference runtimes, feature stores, and orchestration frameworks, while assessing build-versus-buy tradeoffs.
  • Optimize GPU utilization, model batching, and cloud costs across training and inference workloads.
  • Architect ML serving infrastructure that supports traffic surges, seasonal spikes, model transitions, monitoring, alerting, and automated recovery.
  • Mentor junior and mid-level engineers through design reviews, code reviews, and pairing sessions.
  • Lead root-cause analyses and drive permanent fixes while promoting disciplined on-call operations and accountability.

Requirements

  • 8+ years of industry experience, including at least 4+ years of focused hands-on experience in ML infrastructure, MLOps, or large-scale data platform engineering.
  • Proven experience designing and building MLOps platforms supporting the full model lifecycle from data ingestion and distributed training through real-time inference and model governance.
  • Deep expertise in cloud-native infrastructure, preferably AWS, plus Kubernetes, EKS, Docker, and Terraform or Pulumi.
  • Hands-on mastery of ML frameworks such as PyTorch, TensorFlow, Kubeflow, or SageMaker.
  • Expertise building feature stores and high-throughput data pipelines using technologies such as Spark, Flink, and Kafka, with an understanding of training-serving skew and data consistency.
  • Expertise in ML CI/CD, model versioning, experiment tracking, and blue-green or canary deployment strategies.
  • Experience optimizing GPU utilization, implementing model batching, and reducing cloud infrastructure costs.
  • Strong operational experience with on-call practices, proactive monitoring, reliability improvement, and root-cause analysis.
  • Ability to work through ambiguity, learn quickly, experiment, and thrive in a rapidly scaling startup environment.

Benefits

  • Base salary range of $218,000-$285,000 plus potential bonus and stock.
  • Mission-driven work focused on sustainable, accessible retail.
  • Employment is contingent on a background check.
  • Interviews may be conducted by phone, in person, Google Meet, or Zoom. Quince provides reasonable accommodations and equal employment opportunities.
Quince

About Quince

1,001-5,000 employees

Quince is a consumer technology platform rebuilding retail from the infrastructure up. The company began with a towel. A hotel-grade towel retailed for roughly $80. The manufacturing partner who made it sold it for about $7. Nothing about the product justified the gap. The gap was the system. Quince was built to replace that system. Its proprietary Manufacturer-to-Consumer (M2C) platform connects customers directly to the manufacturing partners behind some of the best-made products in the world, with demand signals flowing back to partners weekly so production follows what customers are actually buying, not a year-old forecast. No promotional calendar, no inventory built on guesses, no layered costs priced into the product. What differentiates Quince is not pricing. It is the system. The proof is in how the company operates: cost breakdowns published on product pages, a 365-day return policy, third-party lab testing, and products removed from the assortment at Quince's cost when they fall below the quality bar. Because the platform produces against real demand, the excess inventory that defines conventional retail is never made. Responsibility is built into the model, not added on top of it. Today, Quince offers curated essentials across 100+ categories, from cashmere and linen to fine jewelry and furniture, at prices typically 50 to 70 percent below comparable market prices, and is one of the fastest-growing consumer platforms in the industry. Visit Quince at quince.com or on Instagram at @onequince.

Contact me