PeopleofSuperintelligence

Researcher

Chitwan Saharia

Chitwan Saharia is a machine learning researcher whose work spans grounded language learning and generative image and video models. He earned a B.Tech. in computer science and engineering at IIT Bombay in 2019 and coauthored BabyAI during his undergraduate years. His research papers include SR3, Palette, Imagen and Imagen Video. Imagen combined large language models with diffusion for text-to-image generation and won a NeurIPS Outstanding Paper award; he co-founded Ideogram in December 2022, whose text-to-image product rendered reliable words inside images in 2023. Since December 2025, he is an AI Research Scientist at Meta.

Now at
Meta
Badges

Timeline

  1. 2018–2019

    BabyAI at Mila

    During his undergraduate studies at IIT Bombay, he interns at Mila in Montreal under Yoshua Bengio and Dzmitry Bahdanau and studies sample efficiency for grounded language learning. He coauthors BabyAI, a platform for that research, with its first arXiv version submitted in 2018 and a revision in 2019.

  2. 2019

    BTech at IIT Bombay

    At IIT Bombay, he completes a B.Tech. in computer science and engineering in 2019.

  3. July 2021

    SR3 at Google Research

    In July 2021, he works as a software engineer on Google Research's Brain Team and coauthors SR3, which adapts denoising diffusion models for image super-resolution through repeated refinement.

  4. 2022

    Palette for image translation

    In 2022, he coauthors Palette, a conditional diffusion framework for image-to-image translation across colorization, inpainting, uncropping and JPEG restoration.

  5. 2022

    Imagen and a NeurIPS award

    He coauthors Imagen in 2022, a text-to-image diffusion model that combines large transformer language models with high-fidelity diffusion models. The paper wins a NeurIPS 2022 Outstanding Paper award.

  6. 2022

    Imagen Video for text-to-video

    In 2022, he coauthors Imagen Video, a text-conditional video system that generates high-definition video through a cascade of diffusion models.

  7. December 2022–October 2025

    Co-founding Ideogram

    He co-founds Ideogram in December 2022 and stays with the company through October 2025. By September 2023, the company's text-to-image product renders reliable words inside images.

  8. Since December 2025

    AI Research Scientist at Meta

    He joins Meta as an AI Research Scientist in December 2025.

Career

  1. Since 2025MetaAI Research Scientist
  2. 2022–2025IdeogramCo-founder
  3. 2021Google ResearchSoftware engineer
  4. Until 2019MilaResearch intern

Education

  1. 2019IIT BombayB.Tech.

From the biography.

  1. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Chitwan Saharia. 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.
  2. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Chitwan Saharia. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.
  3. 16% of AI career moves in the 2020s went from a university to a companyIncludes Chitwan Saharia. 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.
  4. 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.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

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

  1. Chitwan Saharia chitwansaharia.github.io. Checked 22 October 2026.
  2. BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning arxiv.org. Checked 22 October 2026.
  3. people | Preethi Jyothi cse.iitb.ac.in. Checked 22 October 2026.
  4. High Fidelity Image Generation Using Diffusion Models research.google. Checked 22 October 2026.
  5. Image Super-Resolution via Iterative Refinement research.google. Checked 22 October 2026.
  6. Palette: Image-to-Image Diffusion Models research.google. Checked 22 October 2026.
  7. Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding proceedings.neurips.cc. Checked 22 October 2026.
  8. Announcing the NeurIPS 2022 Awards blog.neurips.cc. Checked 22 October 2026.
  9. Imagen Video: High Definition Video Generation with Diffusion Models arxiv.org. Checked 22 October 2026.
  10. Find Chitwan Saharia's email & phone number - AI Research Scientist at Meta, San Francisco Bay Area | SignalHire signalhire.com. Checked 22 October 2026.
  11. Ideogram.ai allensthoughts.com. Checked 22 October 2026.