Research Scientist Intern

About the Role

Meta Reality Labs invents the future of visual computing, developing technologies for immersive visual experiences across Meta's products and AR/VR platforms. Our work integrates artificial intelligence, computer vision, multimodal language models, and maps, focusing on areas like localization, vision-language models, scene text and landmark understanding, and spatial reasoning. These advancements aim to enable devices like AI glasses to understand user location.

As a Research Scientist Intern, you will contribute to the advancement of a fast, lightweight VLM-driven visual localizer. This system reasons over semantic and geometric evidence from images and structured map representations to estimate image location, including the place, floor, and camera's position and heading. You will collaborate with researchers and engineers to develop new models, representations, datasets, and evaluation methods, with opportunities to publish your work and influence future products.

Key Responsibilities

• Research, develop, and prototype advanced technologies in Computer Vision, Multimodal Machine Learning, and Visual Localization. This includes creating compact vision-language models that extract structured spatial evidence from images and localize them against structured map representations of indoor and outdoor environments.
• Deliver research results that are publishable and have a direct impact on Reality Labs or Meta’s Family of Apps.
• Build datasets, metrics, and benchmarks for localization accuracy, latency, and cost, and compare against frontier vision-language models.
• Analyze the efficiency, accuracy, scalability, and stability of deployed technologies to identify impact opportunities through novel research.
• Collaborate with team members throughout the project lifecycle, from research to prototyping to product deployment.

Requirements & Tech Stack

• Experience in one or more of the following areas: Computer Vision, Machine Learning, Vision-Language Models, Visual Localization.
• Proven track record of achieving significant results, demonstrated by grants, fellowships, patents, or co-authored publications at top-tier conferences (e.g., CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML).
• Experience with modern C++ and/or Python.
• Experience with deep learning frameworks such as PyTorch.

Qualifications

• Currently pursuing a PhD degree in computer science, electrical engineering, or related fields.
• Intent to return to degree program after internship completion.
• Interpersonal experience, including cross-group and cross-culture collaboration.
• Strong communication skills and ability to convey research ideas to a broad audience.
• Research experience in visual localization, visual place recognition, image geolocalization, scene text recognition, or camera pose estimation.
• Mathematical background in linear algebra, numerical optimization, and probabilistic estimation, with an understanding of 3D geometry and camera models.
• Experience training or fine-tuning vision-language models, including distillation, reinforcement learning, or efficient inference (e.g., quantization or structured outputs).
• Familiarity with map and place data, floor plans, or positioning signals such as GPS and Wi-Fi.
• Relevant research or software engineering work experiences, including internships, coding competitions, or widely used contributions in open-source repositories.