4 months ago
Base Salary
$201k - $254k/yr
Responsibilities
- Lead research and development of pCTR and conversion prediction models for Instacart’s ads surfaces.
- Develop debiasing and calibration techniques including Mixed Negative Sampling, Inverse Propensity Weighting, counterfactual risk minimization, Platt scaling, and isotonic regression.
- Contribute to Multi-Domain Multi-Task model architecture using Mixture-of-Experts, Transformer sequence modeling, and LoRA adapters.
- Advance TIGER generative retrieval, Semantic ID representation learning, and sequence modeling for ads placements.
- Collaborate on foundation-model initiatives based on autoregressive user behavior prediction.
- Translate business observations into well-defined machine learning research directions with clear evaluation criteria.
- Publish and present findings internally and contribute through design reviews, paper sharing, and experiment retrospectives.
Requirements
- PhD or Master’s degree in machine learning, statistics, computer science, information retrieval, or a closely related quantitative field.
- 6+ years of combined academic and industry experience, including PhD research, applying machine learning to ranking, recommendation, or prediction problems at scale.
- Deep understanding of CTR and conversion prediction modeling, including Deep & Wide, DeepFM, DCN, and multi-task learning architectures.
- Strong foundation in causal inference, counterfactual reasoning, and mitigation of selection bias, position bias, and related training-data biases.
- Proficiency in Python, PyTorch, TensorFlow, JAX, SQL, Spark, and Pandas.
- Experience formulating ambiguous problems into well-scoped machine learning research directions and delivering results through rigorous experimentation.
- Strong written and verbal communication skills for explaining modeling decisions to cross-functional stakeholders.
- Preferred experience with ads ranking, auction-based systems, pCTR, bid optimization, ROAS feedback loops, or marketplace dynamics.
- Preferred experience with autoregressive sequence models, generative retrieval, Transformer-based ranking, Semantic IDs, product embeddings, transfer learning, domain adaptation, LoRA, or adapter-based fine-tuning.
- Publication record in venues such as KDD, WWW, RecSys, NeurIPS, ICML, or SIGIR is preferred.
- Experience mentoring junior engineers or shaping technical direction for a modeling team is preferred.
- Familiarity with LLM-driven recommendation, prompt-based personalization, or AI-assisted model development is preferred.
Benefits
- Flex First remote work arrangement with flexibility to work from home, an office, or another preferred location.
- Regular in-person events to maintain connection and community.
- Market-competitive compensation and benefits based on work location.
- Eligible for a new-hire equity grant and annual refresh grants.
Tech Stack
Categories
About Instacart
Instacart builds a grocery technology platform used by retailers and consumers for online shopping, delivery, and pickup. Its business spans a consumer marketplace, enterprise e-commerce and fulfillment software, and a retail media ads network, generating revenue from fees, subscriptions, software, and advertising. Founded in 2012 and headquartered in San Francisco, the company is publicly traded on Nasdaq and operates across the U.S. and Canada.