PeopleofSuperintelligence

Researcher

Kalyan Vasudev Alwala

Kalyan Vasudev Alwala is a computer vision and multimodal AI researcher who has contributed to Meta's PyTorchVideo, ImageBind, SAM 2 and SAM 3. He completed a Master of Science in Robotics at Carnegie Mellon's Robotics Institute in August 2018 under Maxim Likhachev and coauthored the PyRobot framework in June 2019. His subsequent computer vision work included the PyTorchVideo library in November 2021, single-view 3D reconstruction across more than 150 object categories at CVPR 2022, and the six-modality ImageBind embedding in May 2023. In August 2024, he coauthored SAM 2 at AI at Meta, FAIR, which received an honorable mention in the ICLR 2025 outstanding paper awards. He is a research engineer at Meta and works at Meta Superintelligence Labs, where he was a core contributor to SAM 3 in November 2025.

Badges

Timeline

  1. 2018

    Robotics master's at Carnegie Mellon

    He completes a Master of Science in Robotics at Carnegie Mellon's Robotics Institute in August 2018, advised by Maxim Likhachev. His thesis, Integrating Model-based Planning with Skill Learning for Mobile Manipulation, develops a search-based planning framework that combines skills learned from demonstrations with an a priori model.

  2. 2019

    The PyRobot framework

    In June 2019, he is a coauthor of PyRobot, an open-source robotics framework for research and benchmarking. PyRobot is a light-weight, high-level interface on top of ROS that provides hardware-independent mid-level APIs to control different robots.

  3. 2021

    PyTorchVideo for video understanding

    In November 2021, he coauthors PyTorchVideo, an open-source deep-learning library for video understanding. The library provides modular components for tasks including classification, detection, self-supervised learning and low-level processing.

  4. 2022

    3D reconstruction across 150 categories

    At CVPR 2022, he coauthors Pre-Train, Self-Train, Distill: A Simple Recipe for Supersizing 3D Reconstruction. Using image collections from standard recognition datasets, the approach learns 3D inference for over 150 object categories.

  5. 2023

    ImageBind across six modalities

    In May 2023, he is a coauthor of ImageBind, which learns a joint embedding across six modalities: images, text, audio, depth, thermal and IMU data. Image-paired data alone is enough to bind the modalities together.

  6. 2024

    SAM 2 at Meta FAIR

    In August 2024, he is among the authors of SAM 2 from AI at Meta, FAIR, a foundation model for promptable visual segmentation in images and videos. The model is a transformer with streaming memory for real-time video processing, trained on the largest video segmentation dataset to date.

  7. 2025

    An ICLR honorable mention

    ICLR 2025 lists SAM 2: Segment Anything in Images and Videos among its outstanding paper honorable mentions, and he is one of the paper's authors.

  8. Since 2025

    Core contributor on SAM 3

    In November 2025, SAM 3: Segment Anything with Concepts appears with him as a research engineer at Meta and a core contributor at Meta Superintelligence Labs. The unified model detects, segments and tracks objects in images and videos from concept prompts such as short noun phrases.

Career

  1. Since 2025Meta Superintelligence LabsCore contributor

Education

  1. 2018Carnegie Mellon UniversityMaster of Science in Robotics

From the biography.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Kalyan Vasudev Alwala. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. 16% of AI career moves in the 2020s went from a university to a companyIncludes Kalyan Vasudev Alwala. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  3. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Meta. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  4. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Meta Superintelligence Labs. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Kalyan Vasudev Alwala yet. Send us a free-licensed photo.

  1. Kalyan Vasudev Alwala - Robotics Institute Carnegie Mellon University ri.cmu.edu. Checked 22 October 2026.
  2. Integrating Model-based Planning with Skill Learning for Mobile Manipulation - Robotics Institute Carnegie Mellon University ri.cmu.edu. Checked 22 October 2026.
  3. PyRobot: An Open-source Robotics Framework for Research and Benchmarking arxiv.org. Checked 22 October 2026.
  4. [2111.09887] PyTorchVideo: A Deep Learning Library for Video Understanding arxiv.org. Checked 22 October 2026.
  5. CVPR 2022 Open Access Repository openaccess.thecvf.com. Checked 22 October 2026.
  6. ImageBind: One Embedding Space To Bind Them All arxiv.org. Checked 22 October 2026.
  7. SAM 2: Segment Anything in Images and Videos arxiv.org. Checked 22 October 2026.
  8. GitHub - facebookresearch/sam2 github.com. Checked 22 October 2026.
  9. Announcing the Outstanding Paper Awards at ICLR 2025 blog.iclr.cc. Checked 22 October 2026.
  10. SAM 3: Segment Anything with Concepts arxiv.org. Checked 22 October 2026.
  11. GitHub - facebookresearch/sam3 github.com. Checked 22 October 2026.
  12. Segment Anything 3: Live Q&A Meta's Haitham Khedr and Kalyan Vasudev Alwala · Zoom · Luma luma.com. Checked 22 October 2026.