O-1

Data Scientists & the O-1 Visa

No PhD required. No lottery either.

Data scientists and machine learning engineers are among the strongest O-1 candidates in tech. Peer-reviewed publications, competition wins, conference presentations, and the growing demand for ML expertise create a natural evidence base. The O-1A path has no lottery and no annual cap, making it significantly more reliable than the H-1B for data professionals.

Why data scientists qualify

The O-1 visa requires evidence of extraordinary ability, which means demonstrating that you have risen to the top of your field. For data scientists, the evidence often comes from work you have already done. You need to meet at least 3 of the 8 regulatory criteria. Here are the strongest ones for your role.

Strongest criteria for data scientists

You need 3 of the 8 criteria. These are the ones most relevant to your background:

01
Scholarly Articles
Peer-reviewed papers published at NeurIPS, ICML, CVPR, AAAI, KDD, or in journals like JMLR and IEEE TPAMI are strong evidence. Preprints with significant citations on arXiv also carry weight when paired with acceptance at workshops or conferences. Document citation counts, h-index, and the selectivity of the publication venues.
02
Original Contributions
Novel algorithms, ML models, datasets, or frameworks you developed that have been adopted by the research community or deployed in production at scale. Kaggle competition wins (gold medals, top placements in featured competitions), published benchmarks, and open-source ML tools with measurable adoption all qualify.
03
Judging the Work of Others
Serving as a reviewer for NeurIPS, ICML, ICLR, AAAI, or other top ML/AI conferences is direct evidence of this criterion. Program committee memberships, area chair roles, and grant review panel participation also count. Document the number of papers reviewed and the selectivity of reviewer selection.
04
High Remuneration
Data scientists and ML engineers command premium compensation. If your total compensation (salary, stock, bonus) is significantly above the median for your role and geographic area, this criterion is straightforward. Use Bureau of Labor Statistics data or industry salary surveys (Levels.fyi, Glassdoor) as the comparison baseline.
05
Membership in Associations
Fellowship or senior membership in ACM, IEEE, or invitation-only research groups. Acceptance into selective ML research programs, residencies (Google Brain, Meta FAIR, DeepMind), or fellowship programs that require demonstrated excellence for admission.

Common evidence

  • •Published papers with citation counts and venue acceptance rates
  • •Kaggle profile showing competition rankings, medals, and Grandmaster or Master status
  • •Conference reviewer invitations and program committee appointment letters
  • •ML models, datasets, or tools released publicly with download and adoption metrics
  • •Patents for ML systems, algorithms, or data processing methods
  • •Advisory letters from professors, research leads, or recognized experts in your ML subfield
Common Misconception

Myth: “I need a PhD to qualify for the O-1.”

There is no education requirement for the O-1 visa. Data scientists with a bachelor's or master's degree regularly qualify based on publications, competition results, open-source contributions, industry impact, and above-market compensation. What matters is evidence of extraordinary ability in your field, not the degree on your wall. Many approved O-1 holders in data science have no doctorate.

O-1 for Data Scientists

No lottery, no annual cap, no PhD required
Publications at top venues are strong evidence
Kaggle wins and ML competition results count
Conference reviewer roles satisfy judging criterion
Start Your O-1 Case →
Frequently Asked Questions

O-1 questions for data scientists

Do Kaggle competition wins count as evidence?
Yes. Kaggle competition wins and top placements in featured competitions are accepted as evidence of awards and recognition in the field. Gold medals, Grandmaster or Master rankings, and high finishes in competitions with thousands of participants demonstrate that your skills are recognized by your peers. Document the number of participants, your placement, and the competition's significance.
Why is the O-1 better than the H-1B for data scientists?
The H-1B is subject to an annual lottery with roughly 25% selection odds and can only be filed once per year. The O-1 has no lottery, no annual cap, and can be filed at any time with premium processing delivering a decision in 15 business days. For data scientists who meet the evidentiary standard, the O-1 provides a faster, more predictable path to U.S. work authorization.
Can I include preprints or workshop papers as evidence?
Preprints on arXiv that have accumulated significant citations can support your case, especially when combined with acceptances at workshops or conferences. Workshop papers at major venues (NeurIPS workshops, ICML workshops) count as published material. The key is demonstrating that your work has been recognized and has influenced your field.

Ready to begin?

Tell us about your situation and we'll map out the best path forward. Most consultations happen within 48 hours.

Get Started →