Staff Machine Learning Engineer, Fulfillment Planning

DoorDash DoorDash · Consumer · San Francisco, CA · 341 Executive Engineering

Staff Machine Learning Engineer to lead the design, development, and deployment of large-scale production ML systems for DoorDash's fulfillment ecosystem, focusing on real-time decisioning, assignment, and fulfillment estimation. The role involves building 0->1 ML systems, influencing technical direction, setting standards, and mentoring engineers, with a vision towards LLM-inspired foundation models for logistics.

What you'd actually do

  1. Own and build foundational ML systems that directly impact delivery quality, cost, and overall logistics efficiency across DoorDash.
  2. Work on challenging, real-world machine learning problems, including real-time assignment, routing, and fulfillment estimation.
  3. Lead 0→1 ML initiatives, defining how machine learning and optimization are applied across fulfillment products.
  4. Influence architecture, strategy, and execution for a Tier-0 service critical to DoorDash’s logistics platform.
  5. Establish best practices for model development, deployment, monitoring, retraining, and governance.

Skills

Required

  • 8+ years of industry experience building and deploying production-scale machine learning systems
  • strong machine learning fundamentals
  • Python
  • modern ML frameworks, especially deep learning frameworks
  • designed, launched, and operated mission-critical ML models or systems in production, including monitoring, retraining, reliability, and governance
  • lead complex technical projects end to end and influence stakeholders across multiple teams or organizations
  • communicate clearly with both technical and non-technical audiences
  • comfortable operating in ambiguous problem spaces and turning 0→1 ideas into production systems
  • built or shipped large-scale ML models for recommendation, ads, marketplace, logistics, or other domains
  • experience with knowledge distillation from large teacher models into efficient production models

Nice to have

  • LLM-inspired foundation model for intelligence across logistics

What the JD emphasized

  • large-scale production ML systems
  • 0→1 ML systems
  • Staff-level scope
  • foundational ML systems
  • challenging, real-world machine learning problems
  • 0→1 ML initiatives
  • Tier-0 logistics services
  • production-scale machine learning systems
  • large-scale production systems
  • mission-critical ML models or systems in production
  • complex technical projects
  • ambiguous problem spaces
  • 0→1 ideas into production systems
  • large-scale ML models
  • knowledge distillation from large teacher models into efficient production models

Other signals

  • building 0->1 ML systems
  • large-scale production ML systems
  • real-time decisioning
  • define architectures
  • set modeling and deployment standards
  • mentor other engineers
  • shape how DoorDash applies machine learning to logistics at scale
  • foundational ML systems
  • challenging, real-world machine learning problems
  • lead 0->1 ML initiatives
  • influence architecture, strategy, and execution
  • establish best practices for model development, deployment, monitoring, retraining, and governance
  • define and lead DoorDash’s cutting-edge AI vision for logistics: an LLM-inspired foundation model for intelligence across logistics
  • mentor other engineers and raise the technical bar for logistics ML across the organization
  • building and deploying production-scale machine learning systems
  • apply them to large-scale production systems
  • designed, launched, and operated mission-critical ML models or systems in production
  • lead complex technical projects end to end and influence stakeholders
  • operating in ambiguous problem spaces and turning 0->1 ideas into production systems
  • built or shipped large-scale ML models for recommendation, ads, marketplace, logistics, or other domains
  • knowledge distillation from large teacher models into efficient production models