PeopleofSuperintelligence

Researcher

Shivalika Singh

Shivalika Singh is a multilingual-AI researcher and engineer whose first-author papers include the Aya Dataset, Global MMLU and The Leaderboard Illusion. A computer science MSc at Christ University in Bangalore from 2017 to 2020 came first, followed by work at BRIDGEi2i Analytics Solutions as a senior software engineer and then as a self-employed open source developer who led a 50-person Aya project team. The Aya Dataset paper of February 2024 lists Singh as first author; a research engineer role at Cohere followed from June 2024 to January 2026, with first-author work on Global MMLU in 2024 and The Leaderboard Illusion in 2025. Since February 2026, Adaption employs Singh as a Member of Technical Staff.

Works
17
Citations
140
h-index
6
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2017–2020

    Computer science MSc in Bangalore

    Christ University in Bangalore hosts Singh's computer science MSc studies from 2017 to 2020.

  2. 2021–2022

    Senior engineer at BRIDGEi2i

    BRIDGEi2i Analytics Solutions in Bengaluru employs Singh as a senior software engineer from April 2021 to April 2022.

  3. 2022–2024

    Open source work and Aya volunteers

    A self-employed open source career runs from May 2022 to May 2024, with contributions to ZenML, Keras blogs and the Hugging Face Transformers library, a place in the Hugging Face fellowship programme, and a 50-person Aya project team that Singh leads.

  4. February 2024

    The Aya Dataset and Aya Model

    The Aya Dataset paper of February 2024 lists Singh as first author among 33 authors; it describes a human-curated instruction-following dataset spanning 65 languages, built with collaborators from 119 countries. The Aya Model paper of February 2024 names Singh among its 17 authors and introduces a generative language model that follows instructions in 101 languages.

  5. June 2024–January 2026

    Research engineer at Cohere Labs

    Cohere employs Singh as a research engineer in Cohere Labs from June 2024 to January 2026. First-author projects there include Global MMLU and The Leaderboard Illusion, and contributions include the Aya Vision and Aya Expanse releases.

  6. December 2024

    Global MMLU across 42 languages

    Global MMLU appears in December 2024 with Singh as first author, a study of cultural and linguistic biases in multilingual evaluation. It finds that 28% of MMLU questions require culturally sensitive knowledge, and the released Global MMLU covers 42 languages.

  7. 2025

    The Leaderboard Illusion

    The Leaderboard Illusion appears in April 2025 with Singh as first author. It finds undisclosed private testing and selective score reporting on Chatbot Arena and unequal data access between proprietary and open-weight models. The work is presented at NeurIPS 2025, and Singh and Yiyang Nan present it at the Open Science Community's Connect Conference 2025.

  8. Since February 2026

    Technical staff at Adaption

    Adaption takes on Singh as a Member of Technical Staff in February 2026, after the research engineer role at Cohere ends in January 2026.

Career

  1. Since 2026Adaption LabsMember of Technical Staff
  2. 2024–2026CohereResearch engineer
  3. 2021–2022BRIDGEi2i Analytics SolutionsSenior software engineer

Education

  1. 2020Christ UniversityComputer science MSc

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model2024 · Annual Meeting of the Association for Computational Linguistics (ACL) · 52 citationsWith Daniel D'souza, Niklas Muennighoff, Phil Blunsom and Sara Hooker
  2. Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning2024 · Annual Meeting of the Association for Computational Linguistics (ACL) · 29 citationsWith Daniel D'souza, Niklas Muennighoff, Sara Hooker and Sebastian Ruder
  3. Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation2025 · Annual Meeting of the Association for Computational Linguistics (ACL) · 13 citationsWith Antoine Bosselut, Beyza Ermis, Sara Hooker and Sebastian Ruder
  4. The Implications Of The Scheduled Tribes and Other Traditional Forest Dwellers (Recognition Of Forest Rights) Act, 20062014 · Indian Forester · 12 citations
  5. The Leaderboard Illusion2025 · neural information processing systems · 6 citationsWith Beyza Ermis, Noah Smith, Sanmi Koyejo, Sara Hooker and Sayash Kapoor

From OpenAlex, as of 21 October 2026.

  1. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Adaption Labs. 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 Adaption Labs. 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. Among employers with 30+ full-time cases, OpenAI leads primary technical role H-1B pay at $310,000, and Amazon, Google and Microsoft file 57% of all casesIncludes Cohere. Across 440,669 certified US H-1B filings from 44 employers (FY2020–FY2026), OpenAI leads primary technical role base pay at $310,000 and xAI leads overall base pay at $290,000.
  4. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Cohere. 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 Shivalika Singh yet. Send us a free-licensed photo.

  1. Shivalika Singh | alphaXiv alphaxiv.org. Checked 20 October 2026.
  2. [2402.06619] Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning arxiv.org. Checked 20 October 2026.
  3. [2402.07827] Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model arxiv.org. Checked 20 October 2026.
  4. [2412.03304] Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation arxiv.org. Checked 20 October 2026.
  5. The Leaderboard Illusion | Cohere cohere.com. Checked 20 October 2026.
  6. [2504.20879] The Leaderboard Illusion arxiv.org. Checked 20 October 2026.
  7. Open Science Community - Connect Conference labscommunity.cohere.com. Checked 20 October 2026.
  8. Shivalika Singh - Google Scholar scholar.google.com. Checked 20 October 2026.