H-1B Visa for Machine Learning Engineers — Guide

What H-1B Classification Requires for Machine Learning Engineers

A machine learning engineer qualifies for H-1B status when the position meets the regulatory definition of a specialty occupation under 8 CFR 214.2(h)(4)(iii)(A). The role must require theoretical and practical application of a body of highly specialized knowledge, and the beneficiary must hold at least a U.S. bachelor's degree or its foreign equivalent in a specific specialty directly related to the position.

USCIS does not classify occupations by job title. Officers evaluate the actual duties described in the Labor Condition Application (Form ETA-9035) and supporting evidence. A machine learning engineer role satisfies the specialty occupation standard when the petition demonstrates that the position requires application of computer science, mathematics, statistics, or a closely related field at the bachelor's level minimum — and that the specific duties demand this level of education as a standard requirement in the industry or as a legitimate requirement imposed by the employer based on the complexity of the position.

The regulatory test has four prongs, and the petition must satisfy at least one: (1) a bachelor's degree or higher in the specific specialty is normally the minimum requirement for entry into the position in the industry, (2) the degree requirement is common to the industry in parallel positions among similar organizations or is so specialized and complex that it can be performed only by someone with the degree, (3) the employer normally requires a degree for the position, or (4) the nature of the duties is so specialized and complex that the knowledge required is usually associated with attainment of a bachelor's or higher degree. Machine learning roles typically qualify under prong one or two when supported by industry documentation, job postings from comparable employers, and expert opinion letters.

The Labor Condition Application Filing Requirement

Before filing Form I-129 (Petition for a Nonimmigrant Worker), the employer must obtain a certified Labor Condition Application from the Department of Labor. The LCA attests to four conditions: the employer will pay the required wage (the higher of the actual wage paid to similarly employed workers or the prevailing wage for the occupation in the area of intended employment), employment of the H-1B worker will not adversely affect the working conditions of U.S. workers similarly employed, there is no strike or lockout at the place of employment, and the employer has provided notice of the LCA filing to the bargaining representative or, if none exists, has posted notice at the worksite.

As of January 2026, employers file the LCA electronically through the Department of Labor's FLAG system at flag.dol.gov. Certification typically processes within seven business days if the application is complete and contains no obvious errors. The certified LCA is valid for the period of employment stated on the form, up to three years, and must remain in effect throughout the H-1B validity period. Employers must maintain the LCA in a public access file available for inspection by DOL or any member of the public.

The prevailing wage determination is a frequent source of confusion. DOL maintains wage data by Standard Occupational Classification (SOC) code and geographic area. Machine learning engineers are commonly classified under SOC 15-1211 (Computer Systems Analysts), 15-1252 (Software Developers), or 15-2051 (Data Scientists), depending on the actual duties. The employer must request a prevailing wage determination from the National Prevailing Wage Center or use an alternative legitimate wage source, such as the Online Wage Library, and must select the wage level (I through IV) that corresponds to the position's requirements and the worker's qualifications. Misclassification of the wage level is a common RFE trigger.

H-1B Cap Registration and Lottery Procedures

Most H-1B petitions are subject to the annual numerical cap: 65,000 visas for beneficiaries with at least a U.S. bachelor's degree or foreign equivalent, plus an additional 20,000 for beneficiaries holding a U.S. master's degree or higher. Cap-exempt employers include institutions of higher education, nonprofit entities related to or affiliated with institutions of higher education, and nonprofit or governmental research organizations.

As of March 2026, USCIS conducts an electronic registration process each fiscal year before accepting cap-subject petitions. Employers (or their authorized representatives) register beneficiaries during a designated registration period, typically in March for the fiscal year beginning October 1. USCIS then conducts a lottery to select registrations up to the numerical limit. Only petitioners whose registrations are selected may file cap-subject H-1B petitions for that fiscal year.

The registration system operates on a beneficiary-centric model: each beneficiary may be registered only once per fiscal year, regardless of how many employers wish to sponsor them. Multiple registrations for the same beneficiary result in rejection of all registrations for that individual. The registration fee is paid per beneficiary, and as of 2026, employers should confirm the current fee on the USCIS fee schedule at uscis.gov/forms before the registration period opens. Selection in the lottery does not guarantee petition approval — it grants eligibility to file, and the petition is then adjudicated on its merits.

Educational Credential Evaluation for Foreign Degrees

When the beneficiary's highest degree was awarded by a foreign institution, the petition must include a credentials evaluation from a qualified evaluator. The evaluation must conclude that the foreign degree is equivalent to a U.S. bachelor's degree or higher in a specific specialty related to the position. USCIS does not maintain a list of approved credential evaluators, but the evaluator must possess the authority and expertise to render such opinions — typically an organization that is a member of the National Association of Credential Evaluation Services (NACES) or the Association of International Credential Evaluators (AICE).

Machine learning roles frequently involve degrees in computer science, electrical engineering, mathematics, statistics, data science, or computational fields. The evaluator must analyze the foreign institution's accreditation, the curriculum, course descriptions, credit hours, and grading standards to determine U.S. equivalency. A three-year bachelor's degree from certain countries may not equate to a U.S. bachelor's without additional education or progressive work experience. The evaluation must be specific: a conclusion that the degree is "equivalent to a U.S. bachelor's degree in engineering" may not satisfy USCIS if the specialty occupation requires computer science specifically.

Some beneficiaries qualify through a combination of education and progressive experience. The regulatory standard permits three years of progressively responsible experience in the specialty to substitute for one year of university-level education, but this equivalency must be documented in the credentials evaluation, and USCIS applies it strictly. A beneficiary with a three-year degree and three years of specialized experience may qualify for a role requiring a four-year degree if the evaluator concludes the combination is equivalent — but the experience must be specialized, progressive, and directly related to the duties.

Comparison of H-1B Filing Routes for Machine Learning Engineers

What If the Beneficiary Has Only a Three-Year Degree?

Let's be direct: a three-year bachelor's degree from a foreign institution does not automatically equate to a U.S. four-year bachelor's degree, and USCIS applies this standard strictly. Whether the degree qualifies depends on the issuing country's education system, the institution's accreditation, and the field of study.

Beneficiaries with three-year degrees from certain countries — particularly India, the United Kingdom, and parts of Europe — face routine RFEs on educational equivalency. The petition must address this proactively with a detailed credentials evaluation explaining why the specific degree is equivalent, often based on the institution's standing, the intensity of study, or the beneficiary's subsequent education. Some beneficiaries strengthen the petition by documenting a postgraduate diploma, additional coursework, or professional certifications that bring the total education to U.S. bachelor's equivalency.

Alternatively, the combination-of-education-and-experience approach permits the credentials evaluator to conclude that a three-year degree plus progressive work experience in the specialty totals the equivalent of a U.S. bachelor's. USCIS scrutinizes this closely: the experience must be in the specialty (not general IT or unrelated work), it must be progressive (increasing responsibility, not repetitive tasks), and it must be thoroughly documented with detailed employer letters describing duties, dates, and the specialized nature of the work. The Law Offices of Peter D. Chu reviews credential evaluations during the $250 initial consultation to identify equivalency gaps before filing.

Here's the Honest Answer: Most RFEs Trace to the Job Description

Here's the honest answer: the petition lives or dies on the specificity of the job description and the evidence tying the duties to the degree requirement. USCIS officers do not accept generic descriptions — "develop machine learning models," "analyze data," "collaborate with teams" — as proof that the role requires a bachelor's degree in a specialized field. The duties must be described at a level of technical detail that makes the degree requirement self-evident.

A strong job description for a machine learning engineer identifies the algorithms and frameworks the position uses (neural networks, reinforcement learning, natural language processing, computer vision), the tools and languages required (Python, TensorFlow, PyTorch, scikit-learn), the type of data handled (structured, unstructured, time-series), the business problem the models address, and the outcome the position is responsible for delivering. It explains what theoretical knowledge from computer science, mathematics, or statistics the role applies — supervised learning theory, optimization, linear algebra, probability, calculus — and why someone without formal training in these areas could not perform the duties competently.

Supporting evidence should include: detailed letters from the petitioning employer explaining how the role fits into the organization's operations, industry reports or labor market data showing that parallel positions require a bachelor's degree as a standard minimum, expert opinion letters from professors or industry professionals in the field, and copies of job postings from comparable employers listing degree requirements for similar roles. The more specialized or cutting-edge the machine learning work, the stronger the case that only someone with advanced training can do it — but that specialization must be documented, not assumed.

What If the Petition Receives an RFE?

A Request for Evidence asks the petitioner to submit additional documentation to establish eligibility. Common RFE topics for machine learning engineer petitions include: proof that the position qualifies as a specialty occupation, evidence that the beneficiary's education is equivalent to a U.S. bachelor's degree in the required field, documentation that the employer has the ability to pay the proffered wage, clarification of the beneficiary's job duties or the employer's business operations, and verification that the LCA was filed correctly.

The RFE specifies a response deadline — typically 84 days from the date of the notice — and states exactly what evidence USCIS requires. The response must address every point raised in the RFE directly and thoroughly. A partial response or one that sidesteps the officer's concerns almost always results in denial. USCIS does not issue a second RFE for the same petition; the response is the petitioner's one opportunity to satisfy the officer's objections.

Responding to an RFE is not simply gathering more documents — it is diagnosing why the officer questioned eligibility in the first place and reframing the evidence to answer that concern. If the RFE challenges the specialty occupation determination, the response might include additional expert letters, labor market surveys, or a more detailed breakdown of how the duties require application of degree-level knowledge. If the RFE questions educational equivalency, the response might commission a second evaluation from a different credentialing agency or document the beneficiary's progressive experience in greater detail. The Law Offices of Peter D. Chu structures RFE responses to meet the specific regulatory standard the officer cited, not to repeat the original argument louder.

What If the Beneficiary Needs to Start Work Before October 1?

Cap-subject H-1B approvals authorize employment starting October 1 of the fiscal year for which the petition was filed. A beneficiary selected in the March lottery whose petition is approved in June cannot begin H-1B employment until October 1. Employers who need the beneficiary to start earlier must use a different status.

The most common bridge is F-1 Optional Practical Training (OPT) or STEM OPT for beneficiaries who recently graduated from U.S. institutions. OPT authorization permits work for the same employer named in the H-1B petition without violating status, and the beneficiary transitions from OPT to H-1B on October 1 without leaving the United States. The employer must ensure the role described in the OPT application aligns with the H-1B job description — USCIS and SEVP sometimes compare the two.

Another option is cap-gap extension, which automatically extends F-1 status and work authorization through October 1 for students whose H-1B petitions are filed while they are on OPT and whose OPT would otherwise expire before the H-1B start date. The cap-gap extension applies only to timely-filed cap-subject petitions; it does not help beneficiaries outside the U.S. or those in other nonimmigrant categories.

Beneficiaries currently in the U.S. in another work-authorized status (L-1, O-1, TN) may continue working for their current employer under that status while the H-1B petition is pending. On approval, they may choose to activate the H-1B status immediately through a change-of-status request included in the petition, or they may maintain their current status until October 1 and activate H-1B at that time. Some beneficiaries prefer to delay activation to preserve the full six-year H-1B clock.

Premium Processing and Standard Processing Timelines

Form I-129 petitions may be filed with premium processing by submitting Form I-907 and paying the premium processing fee. As of 2026, confirm the current premium processing fee on the USCIS fee schedule before filing. Premium processing guarantees a response (approval, denial, RFE, or notice of intent to deny) within 15 calendar days of USCIS receiving the request. If USCIS does not respond within the guaranteed window, the premium fee is refunded, but processing continues.

Premium processing does not guarantee approval — it guarantees a faster decision. A petition with evidentiary deficiencies receives an RFE just as quickly as an approvable petition receives approval. Premium processing is most valuable when the beneficiary's start date is approaching, when the employer needs certainty for business planning, or when the petition is straightforward and unlikely to generate an RFE.

Regular processing times vary by service center and petition type. USCIS publishes posted processing times at uscis.gov, but these are estimates based on current workload, not guarantees. Processing times for H-1B petitions fluctuate throughout the year — petitions filed immediately after the cap lottery often process faster than petitions filed later in the fiscal year when service centers are handling a broader mix of cases. Employers should check current processing times before deciding whether to pay for premium processing.

Maintaining H-1B Status and Extensions

H-1B status is granted in increments up to three years and may be extended for a maximum of six years total. Beneficiaries who have not used the full six years may file for extensions in one- to three-year increments, provided the employer files a new LCA and Form I-129 before the current status expires. The petition must demonstrate continued eligibility: the position still qualifies as a specialty occupation, the beneficiary still holds the required credentials, and the employer will pay the required wage.

Beneficiaries may qualify for extensions beyond six years under certain conditions tied to the employment-based green card process. If a labor certification (PERM) or Form I-140 immigrant petition has been pending for 365 days or more, the beneficiary may extend H-1B status in one-year increments beyond the six-year limit under AC21 § 106(a). If an approved Form I-140 exists but a visa number is not yet available due to per-country backlogs, the beneficiary may extend H-1B status in three-year increments under AC21 § 106(b). Machine learning engineers from India and China frequently use these provisions due to the severe EB-2 and EB-3 backlogs for those countries.

H-1B workers must maintain status by working only for the petitioning employer in the approved role and location (unless an amended petition is filed and approved), notifying USCIS of any material changes, and not exceeding the period of authorized stay. Termination of employment terminates H-1B status immediately, though the beneficiary has a 60-day grace period (or the time remaining on the I-94, whichever is shorter) to depart the U.S., change status, or find a new H-1B sponsor who files a transfer petition. The grace period does not authorize employment — only lawful presence while arranging departure or new status.

Path from H-1B to Permanent Residence

H-1B status is classified as dual intent, meaning the beneficiary may pursue permanent residence (a green card) without jeopardizing H-1B status. Many machine learning engineers sponsored for H-1B ultimately pursue employment-based green cards in the EB-2 or EB-3 preference categories.

The employment-based green card process has three stages: labor certification (PERM), immigrant petition (Form I-140), and adjustment of status or consular processing (Form I-485 or immigrant visa application). The PERM process requires the employer to test the U.S. labor market and obtain certification from the Department of Labor that no able, willing, qualified U.S. worker is available for the position. This process alone can take six to twelve months.

After PERM approval, the employer files Form I-140, which establishes that the position requires at least a bachelor's degree (EB-3) or an advanced degree or exceptional ability (EB-2), and that the beneficiary meets those requirements. Machine learning engineers with master's degrees commonly qualify for EB-2, which has a faster-moving priority date queue than EB-3 for most countries — but not for India or China, where both categories face multi-year backlogs.

Once the I-140 is approved and a visa number becomes available (based on the priority date and the monthly Visa Bulletin published by the Department of State), the beneficiary may apply for adjustment of status if in the U.S., or process through a U.S. consulate abroad. During the wait for a visa number, the beneficiary remains in H-1B status (extended beyond six years as described above). Spouses in H-4 status may apply for work authorization if the principal beneficiary has an approved I-140, though this policy has been subject to litigation and regulatory changes — confirm the current rule before relying on H-4 EAD.

Consultation and Case Preparation

Immigration petitions are individual determinations based on the specific facts of the employer, the position, and the beneficiary. The material in this article describes the general regulatory framework — not legal advice applying to any particular case. Outcomes depend on how the evidence addresses the regulatory standards, how USCIS interprets the petition under current adjudication policies, and whether the beneficiary's credentials and the employer's business operations satisfy the specialty occupation test as applied by the reviewing officer.

The Law Offices of Peter D. Chu, located at 4615 Convoy St, San Diego, CA 92111, offers initial consultations for employment-based nonimmigrant visa matters. The consultation fee is $250. During the consultation, the firm reviews the beneficiary's educational credentials, the employer's job description and business documentation, and the evidentiary standards the petition must meet. The firm handles H-1B petitions for technology workers, including machine learning engineers, data scientists, software developers, and other specialty occupation roles, and assists employers and beneficiaries through the LCA process, cap registration, petition preparation, RFE response, and status maintenance. The firm's attorneys are familiar with the technical and regulatory nuances of computer science and engineering roles and the common adjudication issues those petitions encounter. Schedule a consultation by calling 858-268-8823 or visiting peterchu.com.

Legal Disclaimer: This article provides general information about H-1B classification for machine learning engineers under U.S. immigration law. It is not legal advice and does not create an attorney-client relationship between the reader and the Law Offices of Peter D. Chu or any of its attorneys. Immigration outcomes depend on the specific facts of each case, and eligibility standards are applied by USCIS on a case-by-case basis. Do not rely on this article as a substitute for consultation with a licensed immigration attorney. For advice on your specific situation, contact an attorney directly.

Schedule a consultation with the Law Offices of Peter D. Chu — 4615 Convoy St, San Diego, CA 92111 · 858-268-8823 · Mon–Fri, 8:30 AM–5:30 PM. Consultation fee: $250.