H1B Visa Guide
Artificial intelligence and machine learning engineers are among the most in-demand H1B workers. Here's your complete roadmap to sponsorship in 2025.
Artificial intelligence and machine learning engineering sits at the intersection of mathematics, statistics, and software engineering — a combination that firmly qualifies as a specialty occupation under H1B regulations. USCIS has consistently approved H1B petitions for AI/ML roles when employers demonstrate that the position requires a theoretical and practical application of highly specialized knowledge.
The typical educational requirement is a bachelor's degree or higher in computer science, data science, electrical engineering, mathematics, or statistics. Employers must document that the degree requirement is standard for the position and that the specific AI/ML duties cannot be performed without that specialized educational background.
One nuance that AI/ML engineers should understand: the precise job title matters less than the actual duties. USCIS reviews the nature of the work, not the label. A "machine learning engineer" performing routine software development may face more scrutiny than an "applied scientist" building neural network architectures, even if the former sounds more technical. Your employer's attorney should carefully craft the LCA and petition to emphasize the specialized, theoretical nature of the work.
AI/ML engineers benefit from strong demand signals in the H1B market. According to DOL LCA data, thousands of H1B positions are certified annually for AI and machine learning roles, with approvals concentrated at major technology firms, financial institutions, and research universities.
The SOC code selection critically affects your prevailing wage obligation and can influence how USCIS categorizes your position. AI/ML engineers most commonly use SOC 15-2051 (Data Scientists), SOC 15-1252 (Software Developers, Applications), or SOC 15-1299 (Computer Occupations, All Other). Some employers use SOC 15-1211 (Computer Systems Analysts) for roles with a systems integration focus.
SOC 15-2051 (Data Scientists) typically carries the highest prevailing wages — Level III wages in San Francisco can exceed $200,000 annually. This code is most appropriate when the role involves building predictive models, running statistical analyses, and interpreting data patterns for business decisions. The AI specialization within data science makes this classification defensible even for roles focused on deep learning or NLP.
SOC 15-1252 (Software Developers) is used when the ML engineer's primary function is software development — building ML pipelines, MLOps infrastructure, or production AI systems. This code has well-established precedent and typically results in smoother LCA certification. Level II wages in Seattle under this code average around $145,000, while Level III approaches $175,000.
Your employer's attorney should choose the SOC code that most accurately reflects the actual job duties, not the code with the lowest prevailing wage. Misclassification is a common source of DOL audits and USCIS RFEs. Consistency between the LCA job description and the H1B petition job duties is critical.
The AI/ML H1B landscape is dominated by major technology companies, but the opportunity extends far beyond FAANG. Google, Meta, Amazon, Microsoft, and Apple collectively file thousands of H1B petitions for AI/ML roles annually. These companies have dedicated immigration teams, established processes, and routinely win lottery selections at high rates due to volume.
AI-native companies have become major sponsors in recent years. OpenAI, Anthropic, Cohere, Stability AI, and dozens of well-funded AI startups actively recruit international talent and sponsor H1B visas. These employers often offer compensation well above prevailing wage — a factor that strengthens the H1B petition and reduces RFE risk.
Financial institutions are an underappreciated source of AI/ML H1B sponsorship. Goldman Sachs, JPMorgan Chase, Two Sigma, Citadel, and D.E. Shaw employ large AI/ML teams and routinely sponsor H1Bs. These firms use AI for algorithmic trading, risk modeling, fraud detection, and customer analytics. Prevailing wages at these firms are typically at Level III or IV.
Universities and national laboratories offer cap-exempt H1B positions — meaning they bypass the annual lottery entirely. If you're an AI/ML researcher affiliated with a university or nonprofit research organization, cap-exempt sponsorship can be filed at any time of year with no lottery risk. Many AI researchers strategically pursue academic positions to secure H1B status before moving to industry.
The H1B process begins with your employer filing a Labor Condition Application (LCA) with the Department of Labor, typically approved within 7 business days. The LCA certifies the job title, SOC code, location, and prevailing wage commitment. For AI/ML roles, employers should ensure the LCA accurately reflects all work locations, including remote work arrangements and client site requirements.
The H1B petition itself is filed with USCIS during the regular cap season (April 1 filing) or under cap-exempt rules if your employer qualifies. With premium processing (Form I-907, $2,805 fee as of 2024), USCIS guarantees a decision within 15 business days. Most AI/ML engineers and their employers opt for premium processing given the importance of the role.
RFEs are relatively common for AI/ML H1B petitions, particularly when the job title is novel or the duties are difficult to map to established SOC codes. Common RFE issues include requests for additional evidence of specialty occupation status, employer-employee relationship documentation for consulting roles, and degree equivalency challenges when the beneficiary holds a degree in a non-CS field.
Maintaining H1B status requires that your job duties, location, and compensation remain consistent with the approved petition. Significant changes — a new role at the same employer, a different work location, or a substantial salary reduction — may require an amended H1B petition. AI/ML engineers should communicate proactively with their employer's immigration attorney when job responsibilities evolve.
AI/ML engineers pursuing permanent residence have several pathways available. The EB-2 category covers professionals holding advanced degrees or those with exceptional ability — both common among senior AI/ML engineers. The EB-2 National Interest Waiver (NIW) is particularly attractive for AI researchers who can demonstrate their work benefits the United States broadly.
EB-1A (Extraordinary Ability) is achievable for AI/ML engineers with substantial publication records, citations, awards, or demonstrated influence on the field. If you've published at top venues like NeurIPS, ICML, or CVPR with significant citations, your immigration attorney should evaluate EB-1A eligibility seriously. EB-1A allows self-petition without employer sponsorship and has no priority date wait for most nationalities.
EB-1B (Outstanding Researcher) requires employer sponsorship and evidence of international recognition for outstanding achievements in a particular academic field. For AI/ML researchers at universities or corporate research labs, EB-1B can be a faster path than the standard PERM labor certification route.
For most AI/ML engineers from India or China, the EB-2 and EB-3 backlogs create multi-year or multi-decade waits. During this period, maintaining valid H1B status (and H1B extensions beyond six years under AC21) is essential. PERM labor certification timing, priority date optimization, and concurrent filing strategy should all be discussed with an experienced immigration attorney early in your career.
H1B Job Board Editorial Team
Immigration Research & Career Intelligence
Our team tracks LCA filings, USCIS approval data, and employer sponsorship trends across technology sectors. All guides are reviewed for accuracy against current DOL and USCIS policy.