Senior Machine Learning Engineer - Search & Recommendations Ranking

Overview

About the Role:

This is a general posting for multiple Sr. Machine Learning roles open across our 4-sided marketplace. You’ll get the chance to learn about the problems the different ML teams solve as you go through the process. Towards the end of your process, we’ll do a team-matching exercise to determine which of the open roles/teams you’ll join. You can find a blurb on each team at the bottom of this page. 

About the Team:

The Search & Recommendations ML team is Instacart’s engine for multi-task, multi-objective ranking—unifying search, discovery, ads, and merchandising into a single value-aware platform. Partnering with world-class engineers, scientists, and PMs, we build the ranking backbone that powers every pixel of the shopping journey - optimizing not just for clicks, but for incremental GTV, basket lift, and retention over the long run.

What we’re building:

• Foundational Ranking Backbone Models -  A set of multi-task / multi-objective models (shared encoders + task heads) that jointly learn relevance, conversion, margin contribution, churn risk, and ad quality, enabling consistent decisions across search and recommendations
• Value-Aware Optimization - Uplift and long-horizon value models to steer decisions toward incrementality and LTV, with calibrated constraints (quality, diversity, fairness, spend pacing) and guardrails for safe exploration.
• LLM-Enhanced Retrieval & Features - LLMs enrich query and item semantics for long-tail recall, generate features for cold-start, and feed the ranker with reasoning-rich context—while ranking remains the source of truth for final ordering.

Our commitment to AI innovation is reflected in our recent publications and research contributions to the field (Recent publications 1, 2, 3, 4, 5, 6). 

About the Job:

• Architect the ranking backbone that unifies query understanding, personalization, multi-objective ranking, ads, and merchandising into a single adaptive platform.
• Design long-horizon objective functions (incrementality, LTV, habit formation) and build uplift/causal value models that move us beyond short-term engagement.
• Develop production-grade MTL (e.g., shared encoders, MMOE/PLE task heads) to jointly learn relevance, propensity, margin, and churn risk—calibrated, constrained, and explainable.
• Own the inference layer: goal-aware re-rankers, diversity and quality constraints, safe exploration, and millisecond-class latency.
• Advance evaluation: online experiments, long-horizon cohort metrics, counterfactual/randombucket evaluation, and attribution pipelines for incremental GTV and retention.
• Partner across ads, infra, product, and design to translate business goals into ranking policies and measurable ROI.
• Mentor ML engineers, growing depth in ranking, causal inference, and large-scale serving

About You:

Minimum Qualifications:

• 5+ years applying ML at scale (3+ in technical leadership), with a track record improving ranking/recommendation systems in production.
• Proven impact using multi-objective or constrained optimization to balance relevance, revenue, margin, and UX; experience with online testing and attribution beyond CTR.
• Strong coding (Python) and data fluency (SQL/Pandas); solid with classic ML (e.g., XGBoost) and deep learning (TensorFlow/PyTorch).
• Excellent analytical skills and cross-functional communication.

Preferred Qualifications:

• Expertise in multi-task learning (MMOE/PLE, shared encoders), calibration, counterfactual evaluation, uplift/causal modeling, and/or contextual bandits for exploration.
• Experience building low-latency ranking services (feature stores, caching, vector + lexical retrieval, re-ranking, A/B infra) and constraint-aware inference.
• Hands-on with LLMs as feature/recall enhancers (embeddings, adapter tuning) with a clear understanding of when the ranker should arbitrate.