Senior Machine Learning Engineer II, Fulfillment, Matching and Positioning

Overview

Instacart’s Logistics organization powers the intelligence and execution behind our fulfillment system. We’re hiring a Senior Machine Learning Engineer to join the Matching & Positioning team, a tight-knit group of 9 engineers and scientists focused on real-time decisioning for order batching, shopper routing, and assignment across a dynamic, multi-sided marketplace.

In this role, you’ll work at the intersection of operations research, combinatorial optimization, and machine learning to design and ship algorithms that directly impact profitability, on-time delivery, shopper experience, and customer satisfaction at scale. You’ll collaborate closely with engineering, product, and data science partners to translate ambiguous problems into well-formed optimization and ML systems that operate under sub-second latency and high throughput.

If you thrive in a fast-paced environment, enjoy rolling up your sleeves, and want to see your models make decisions in the real world every minute of every day, this team is for you.

About the Job

You will build production-grade optimization and ML solutions that drive Instacart’s fulfillment decisions end-to-end in a rapidly evolving, high-scale environment.

• Design, implement, and deploy algorithms for order batching, real-time shopper assignment, routing, and marketplace positioning using techniques such as MIP/CP-SAT, heuristics/metaheuristics, and learning-to-rank.
• Own the full model lifecycle: problem formulation, data pipelines and features, offline evaluation and simulation, A/B testing, staged rollouts, and ongoing monitoring/observability.
• Build reliable, low-latency services in Python (and, where performance dictates, C++ or Go) that integrate with solvers (e.g., OR-Tools, Gurobi, CPLEX) and run on cloud infrastructure with Docker/Kubernetes.
• Partner with product, operations, and data science to define roadmaps and success metrics; deliver measurable impact to on-time rates, shopper utilization, cost per order, and customer experience.
• Leverage experimentation and causal methods along with offline counterfactual replay/simulation to validate changes and de-risk launches.
• Contribute to engineering excellence through code reviews, design docs, robust testing, and participation in an on-call rotation for mission-critical fulfillment services; mentor peers and raise the technical bar.

This is a fast-moving domain with evolving constraints and objectives. Success requires comfort with ambiguity, pragmatic prioritization, and a bias toward iterative learning and continuous improvement.

About You

You pair a deep toolkit in operations research and machine learning with strong software engineering fundamentals. You’re motivated by real-world impact, communicate clearly with cross-functional partners, and take ownership from ideation to production.

Minimum Qualifications

• Bachelor’s degree in Computer Science, Operations Research, Electrical Engineering, Applied Mathematics, or a related field (or equivalent practical experience).
• 5+ years of professional experience building and shipping ML and/or optimization systems to production.
• 3+ years formulating and solving large-scale combinatorial optimization problems (e.g., VRP, matching, scheduling) using solvers such as OR-Tools, Gurobi, or CPLEX (MIP/CP-SAT) and heuristic methods.
• Proficiency in Python and SQL, including writing production-quality code with testing, profiling, and code review practices.
• Hands-on experience deploying algorithms/models as microservices with Docker and Kubernetes on a major cloud provider (GCP or AWS), including monitoring, alerting, and dashboards.
• Experience designing and operating low-latency decision services in high-throughput environments (targeting sub-second P95 response times).
• Practical experience with A/B testing or online experimentation platforms, from hypothesis through analysis and rollout decisions.
• Strong collaboration and communication skills with engineering, product, and data science stakeholders.

Preferred Qualifications

• Master’s or PhD in Operations Research, Computer Science, Electrical Engineering, Applied Mathematics, or a related quantitative field.
• Domain experience in logistics, ride-hailing, delivery, or marketplace optimization at scale.
• Familiarity with reinforcement learning or contextual bandits for online decision-making and exploration/exploitation tradeoffs.
• Experience with geospatial data, routing APIs, and graph algorithms.
• Background in building simulation frameworks and counterfactual evaluation for decision systems.
• Experience with streaming data and real-time feature computation (e.g., Kafka, Flink) and feature stores.
• Proficiency in C++ or Go for performance-critical components.
• Track record of mentoring engineers and leading cross-functional projects to measurable outcomes.
• Experience participating in an on-call rotation for production ML/optimization services.

#LI-Remote