3d Computer Vision Research Engineer, Apple Maps

Apple Apple · Big Tech · Cupertino, CA +1 · Machine Learning and AI

Research Engineer for Apple Maps focusing on 3D computer vision and ML to extract map features from imagery and sensor data. Develops models for visual understanding, scene understanding, and physical-world reasoning, combining ML, 3D geometry, and multimodal data for large-scale mapping problems.

What you'd actually do

  1. Prototype, develop, and evaluate computer vision and machine learning algorithms for real-world map feature extraction problems.
  2. Build models and systems for visual understanding, scene understanding, map automation, and physical-world reasoning.
  3. Work with large multimodal datasets, including aerial, satellite, and ground imagery.
  4. Develop methods for detecting, extracting, classifying, and validating map-relevant features from imagery and sensor data.
  5. Collaborate with cross-functional teams to bring research ideas into production-quality systems.

Skills

Required

  • Computer Vision
  • Machine Learning
  • Object Detection
  • Segmentation
  • Feature Extraction
  • Scene Understanding
  • Python
  • C++

Nice to have

  • 3D Geometry
  • Spatial Reasoning
  • Geospatial Data
  • Vision-Language Models
  • Multimodal LLMs
  • Reasoning Models
  • Generative Image/Video Models
  • Computer Graphics

What the JD emphasized

  • Strong background in computer vision and machine learning
  • Hands-on experience with visual understanding, object detection, segmentation, feature extraction, scene understanding, or related computer vision problems
  • Familiarity with 3D geometry, spatial reasoning, or large-scale geospatial data is a plus
  • Publications in top-tier CV/ML conferences, such as CVPR, ICCV, ECCV, NeurIPS, ICLR, SIGGRAPH, or related venues
  • Experience with vision-language models, multimodal LLMs, reasoning models, or generative image/video models is a plus
  • Experience building ML systems for large-scale real-world visual data
  • Knowledge of 3D geometry, geospatial data, or computer graphics fundamentals is a plus

Other signals

  • develops models and systems that extract meaningful map features from real-world imagery
  • develop novel methods that combine machine learning, 3D geometry, multimodal data, and large-scale systems
  • visual understanding, object detection, segmentation, feature extraction, scene understanding