Engineering Manager, Machine Learning (Caper)

Instacart is transforming the grocery industry by providing innovative solutions such as AI-powered shopping carts. They are seeking an Engineering Manager, Machine Learning to lead a team of engineers in developing perception and reasoning systems for these carts, focusing on enhancing item recognition, checkout speed, and system reliability.

Responsibilities:

• Lead and grow a team of ~10 ML and AI infrastructure engineers building the perception and reasoning systems that power Caper Carts in live retail environments
• Define the technical vision, roadmap, and success metrics for cart perception and multimodal understanding; prioritize work that drives measurable gains in item recognition accuracy, checkout speed, and system reliability
• Architect scalable training, data, and inference platforms on GCP using Ray, Kubernetes, and modern MLOps practices to enable rapid experimentation and safe, repeatable deployments
• Deliver production-grade CV/VLM models for multi-camera item detection, weighing, and basket reasoning; optimize on-device inference for low-latency, high-availability operation at the edge
• Build the data flywheel end-to-end—instrumentation, labeling, evaluation, offline/online testing, and monitoring—to continuously improve performance across diverse store conditions
• Collaborate cross-functionally with Android, hardware, product, design, operations, and retailer partners; communicate risks, tradeoffs, and timelines clearly in a fast-paced, ever-evolving environment

Requirements:

• 8+ years of experience building and deploying machine learning systems, with a strong focus on computer vision in production environments
• 2+ years of experience managing teams of 6+ ML/AI engineers, including hiring, performance management, and career development
• Hands-on expertise with deep learning (e.g., PyTorch), model training/evaluation, and MLOps practices for reliable CI/CD of ML services
• Proven experience architecting and operating ML infrastructure on GCP (e.g., GKE, Vertex AI, BigQuery) and distributed training/inference with Ray; containerization with Docker and orchestration with Kubernetes
• Experience delivering real-time edge inference, including model optimization (e.g., TensorRT, ONNX, quantization) and monitoring for latency, throughput, and accuracy
• Proficiency in Python and SQL, with a track record of shipping end-to-end CV systems including data pipelines, experimentation, deployment, and post-launch iteration
• Bachelor's degree in Computer Science, Electrical/Computer Engineering, or a related technical field, or equivalent practical experience
• Experience integrating on-device ML with Android applications and collaborating closely with Android teams on SDKs and APIs
• Background with multimodal vision-language models (VLMs) and large language models (LLMs) for perception, retrieval, or instruction-based reasoning
• Experience with sensors and hardware integration (e.g., multi-camera setups, weight sensors), calibration, and dataset generation for robotics or retail environments
• Demonstrated success leading cross-functional programs across 3+ partner teams and delivering multi-quarter roadmaps
• Graduate degree (MS/PhD) in a relevant field with research or applied focus in computer vision, machine learning, or robotics