PeopleofSuperintelligence

Researcher

Mitchell Wortsman

Mitchell Wortsman is a machine learning researcher whose work covers robust neural networks, open tools for vision and language, and scaling large language models. At AI2, he served as a predoctoral young investigator in 2018–2019 and pursued a PhD at the University of Washington; his 2020 CVPR paper showed that neural networks with random weights can contain useful subnetworks. His 2021 work included OpenCLIP, and he coauthored Model soups and LAION-5B in 2022; LAION-5B won a NeurIPS Outstanding Paper award. He earned a Google PhD Fellowship in 2023 for machine learning research and coauthored a 2025 ICLR study of language-model scaling. At Anthropic, he is a member of technical staff on the pretraining team.

Now at
Anthropic

Timeline

  1. 2018–2020

    AI2 and neural subnetworks

    In 2018–2019, he serves as a predoctoral young investigator at AI2 before pursuing a PhD at the University of Washington. In 2020, he coauthors a CVPR paper showing that neural networks with random weights can contain smaller subnetworks that match a trained ResNet-34 on ImageNet.

  2. 2021

    Building OpenCLIP

    In 2021, he coauthors OpenCLIP, an open-source implementation of CLIP.

  3. 2022

    Model soups and LAION-5B

    In 2022, he coauthors Model soups, which averages fine-tuned model weights to improve accuracy and robustness. He also coauthors LAION-5B, an open dataset of 5.85 billion CLIP-filtered image-text pairs; it receives a NeurIPS Outstanding Paper award.

  4. 2023

    A Google PhD Fellowship

    In 2023, he earns a Google PhD Fellowship for large-scale machine-learning research.

  5. 2025

    Language-model scaling

    In 2025, he coauthors an ICLR study that fits scaling laws with a testbed of 104 language models from 0.011 billion to 6.9 billion parameters.

  6. As of October 2026

    Pretraining at Anthropic

    As of October 2026, he is a member of technical staff on Anthropic's pretraining team.

Career

  1. NowAnthropicMember of technical staff
  2. 2018–2019Allen Institute for AI (Ai2)Predoctoral young investigator

Education

  1. University of WashingtonPhD

From the biography.

  1. 22 of 133 people shaping AI with a former OpenAI role now work at AnthropicIncludes Anthropic. Our estimates of where OpenAI recruits came from and where 133 people with former OpenAI roles work now, including the companies and labs they founded.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Anthropic. 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.
  3. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Anthropic. 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.
  4. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Anthropic. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.

Colleagues

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

Sources and credits

Photo: We have no published photo of Mitchell Wortsman yet. Send us a free-licensed photo.

  1. Roozbeh Mottaghi's Webpage roozbehm.info. Checked 21 October 2026.
  2. Allen School’s Miranda Wei and Mitchell Wortsman earn Google Ph.D. Fellowships for advancing user security and privacy and large-scale machine learning research news.cs.washington.edu. Checked 21 October 2026.
  3. What's Hidden in a Randomly Weighted Neural Network? openaccess.thecvf.com. Checked 21 October 2026.
  4. Mitchell Wortsman mitchellnw.github.io. Checked 21 October 2026.
  5. Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time arxiv.org. Checked 21 October 2026.
  6. LAION-5B: An open large-scale dataset for training next generation image-text models proceedings.neurips.cc. Checked 21 October 2026.
  7. NeurIPS 2022 Awards neurips.cc. Checked 21 October 2026.
  8. Language models scale reliably with over-training and on downstream tasks proceedings.iclr.cc. Checked 21 October 2026.