高级机器学习工程师，广告响应预测
Senior Machine Learning Engineer, Ads Response Prediction
我们正在改变零售行业
在Instacart，我们邀请世界通过食物传递爱，因为我们相信每个人都能获得他们喜爱的食物，并有更多时间与所爱之人一起享受。当其他人看到的只是简单的送货需求时，我们看到了令人兴奋的复杂性和无限的机会，以满足我们社区的各种需求。我们致力于提供客户依赖的必需服务，让他们获取杂货和家庭用品，同时为Instacart个人购物者提供安全且灵活的收入机会。
Instacart已成为数百万人的生命线，我们正在组建团队推动我们的购物车向前发展。如果你准备好做出一生中最好的工作，来加入我们的行列吧。
Instacart是一个以灵活性优先的团队
没有一种方法适用于所有人如何做好工作。我们的员工可以自由选择在何处发挥最佳表现——无论是在家、办公室，还是最喜欢的咖啡馆——同时通过定期的线下活动保持联系并建立社区。了解更多关于我们在工作地点上的灵活方式。
简介
作为Ads Response Prediction团队的高级机器学习工程师，你将负责并执行开发支持Instacart广告生态系统的机器学习模型。这是一个以研究为导向的角色，专注于理论问题建模、训练方法和模型质量，而不是基础设施或全栈工程。你将解决pCTR建模中的重要挑战，如减轻训练数据中的选择偏差、位置偏差和优化器诅咒，提高跨表面和领域的模型校准，以及推进我们的多任务学习和序列建模能力。你将为广告排序的基础模型方法做出贡献，并参与前沿的检索系统，如TIGER（用于生成推荐的Transformer索引）、语义ID和领域语言模型。
Ads Response Prediction团队负责所有系统、算法和机器学习模型，以确保Instacart支持的所有平台的客户获得相关且吸引人的广告体验。这包括搜索和探索检索系统、用于下一次互动推荐的序列建模和生成检索系统、大语言模型集成、相关性模型、pCTR模型、竞价模型和增量模型。该团队优化市场效率，以确保客户愉悦的购物体验，以及理想的广告展示。
查看英文原文
We're transforming the grocery industry
At Instacart, we invite the world to share love through food because we believe everyone should have access to the food they love and more time to enjoy it together. Where others see a simple need for grocery delivery, we see exciting complexity and endless opportunity to serve the varied needs of our community. We work to deliver an essential service that customers rely on to get their groceries and household goods, while also offering safe and flexible earnings opportunities to Instacart Personal Shoppers.
Instacart has become a lifeline for millions of people, and we’re building the team to help push our shopping cart forward. If you’re ready to do the best work of your life, come join our table.
Instacart is a Flex First team
There’s no one-size fits all approach to how we do our best work. Our employees have the flexibility to choose where they do their best work—whether it’s from home, an office, or your favorite coffee shop—while staying connected and building community through regular in-person events. Learn more about our flexible approach to where we work.
Overview
As a Senior Machine Learning Engineer on the Ads Response Prediction team, you will own and execute the development of ML models that power Instacart's ads ecosystem. This is a research-leaning role focused on theoretical problem formulation, training methodology, and model quality rather than infrastructure or full-stack engineering. You will work on meaningful challenges in pCTR modeling such as mitigating selection bias, position bias, and optimizer's curse in training data, improving model calibration across surfaces and domains, and advancing our multi-task learning and sequence modeling capabilities. You will contribute to our foundation model approach for ads ranking and work on cutting-edge retrieval systems like TIGER (Transformer Index for Generative Recommenders), Semantic ID and domain language models.
The Ads Response Prediction team owns all systems, algorithms and ML models to ensure a relevant and engaging Ads experience to customers of all the platforms powered by Instacart. This includes search and exploration retrieval systems, sequential modeling and generative retrieval systems for next interaction recommendations, LLM integrations, relevance models, pCTR models, bidding models and incrementality models. The team optimizes for an efficient marketplace to ensure delightful customer shopping experience, desirable advertiser business outcome and Instacart Ads revenue.
The team has strong ML infrastructure and MLOps support, including Delta/DBT-Spark data pipelines, Ray-based distributed training, and automated model deployment. This means you can focus your energy on advancing modeling science rather than building infrastructure.
About the Job
- Own and execute research and development of pCTR and conversion prediction models, with a focus on improving calibration, reducing training data biases (selection bias, position bias, optimizer's curse), and advancing model accuracy across Instacart's ads surfaces.
- Design and implement debiasing techniques such as Mixed Negative Sampling (MNS), Inverse Propensity Weighting (IPW), counterfactual risk minimization, and calibration methods (Platt scaling, isotonic regression) to address systematic prediction biases.
- Contribute to the next-generation Multi-Domain Multi-Task (MDMT) model architecture, incorporating innovations like Mixture-of-Experts (MoE), Transformer layers for sequential user behavior, and LoRA adapters for scalable domain fine-tuning.
- Contribute to sequence modeling initiatives including the TIGER generative retrieval system and Semantic ID representation learning, expanding their application across ads surfaces such as Product Details, Search and other placements.
- Collaborate with the broader ML community in the company on the path toward Foundation Models using autoregressive user behavior prediction.
- Formulate and scope ambiguous modeling problems within your project scope from first principles. Translate business observations (e.g., overcalibration patterns, cold-start underperformance) into well-defined ML research directions with clear evaluation criteria.
- Publish and present findings internally. Contribute to the team's culture of technical rigor through design reviews, paper sharing, and experiment retrospectives.
About You
Minimum Qualifications
- Master's or PhD in machine learning, statistics, computer science, information retrieval, or a closely related quantitative field; or equivalent experience.
- 3+ years of combined academic and industry experience (including PhD research) applying ML to ranking, recommendation, or prediction problems at scale.
- Deep understanding of CTR/conversion prediction modeling, including familiarity with architectures such as Deep & Wide, DeepFM, DCN, and multi-task learning formulations.
- Strong foundation in causal inference, counterfactual reasoning, and training data bias mitigation. Ability to reason about selection bias, position bias, and propensity-based correction methods.
- Proficiency in Python and deep learning frameworks (PyTorch, Tensorflow, JAX). Fluency in data manipulation tools (SQL, Spark, Pandas).
- Track record of formulating ambiguous problems into well-scoped ML research directions and delivering results through rigorous experimentation.
- Strong written and verbal communication skills. Ability to explain complex modeling decisions to cross-functional stakeholders including product managers and data scientists.
Preferred Qualifications
- Experience in ads ranking or auction-based systems (pCTR, bid optimization, ROAS feedback loops, marketplace dynamics).
- Hands-on experience with autoregressive sequence models for user behavior prediction, generative retrieval, or transformer-based ranking architectures.
- Familiarity with learned representations such as Semantic IDs, product embeddings, or other approaches to reducing feature cardinality and cold-start challenges.
- Experience with transfer learning or domain adaptation techniques (e.g., LoRA, adapter-based fine-tuning) applied to recommendation or ranking models.
- Publication record in top-tier venues (KDD, WWW, RecSys, NeurIPS, ICML, SIGIR, or similar).
- Familiarity with LLM-driven approaches to recommendation, including prompt-based personalization and AI-assisted model development (AutoML).
#LI-Remote
Instacart provides highly market-competitive compensation and benefits in each location where our employees work. This role is remote and the base pay range for a successful candidate is dependent on their permanent work location. Please review our Flex First remote work policy here. Currently, we are only hiring in the following provinces: Ontario, Alberta, British Columbia, and Nova Scotia.
Offers may vary based on many factors, such as candidate experience and skills required for the role. Additionally, this role is eligible for a new hire equity grant as well as annual refresh grants. Please read more about our benefits offerings here.
For Canadian based candidates, the base pay ranges for a successful candidate are listed below.
CAN
$180,000—$190,000 CAD