The most reliable list of companies that sponsor H1B workers is the one the government publishes. USCIS runs the H-1B Employer Data Hub, which covers fiscal year 2009 through fiscal year 2026 (quarter 3) and counts, employer by employer, the H-1B workers USCIS approved and denied. Third-party "top sponsors" lists rank employers too, but you cannot check how they count. Building your own list from the source is not hard, and it answers the question that actually matters to you: which employers in your field, in your region, have recently brought on someone in your situation.
This post covers what the data can and cannot tell you, how to filter it, the difference between cap-subject and cap-exempt employers, the lottery numbers behind all of it, and how to turn a spreadsheet into applications. It is general information, not legal advice. For anything about your own status, talk to your school's designated school official or an immigration attorney.
What sponsorship data can and cannot tell you
A past approval is evidence of two things. The employer was willing to sponsor at least once, and it has a working immigration process: a lawyer, a budget and a person who signs the paperwork. That is worth a lot. Plenty of employers have never done it, and a recruiter at one of them cannot conjure the process for you.
It is not evidence of much else. USCIS's own explanation of the Data Hub lists the limits, and they matter for how you read it:
- First decisions only. "Subsequent decisions, such as on an appeal or revocation, are excluded." Pending petitions are not included either.
- Grouped by when USCIS decided, not when the employer filed.
- The address is a mailing address. City, state and ZIP are the employer's mailing address, which "is not necessarily the beneficiary(ies) work location." A company headquartered in one state can employ its H-1B workers in another.
- Names are messy. Forms are entered by hand, and "data entry and petitioner errors may result in a missing and/or erroneous employer name." USCIS uses "the most common spelling of employer name per unique tax ID," and shows only the last four digits of that ID.
- One employer can be several rows. An employer with multiple addresses in a year has multiple rows.
- Unknown industry is 99. A blank industry code is also assigned 99.
And some things no government table can tell you: whether this team sponsors this year, whether it sponsors at your level, or whether a particular role is open to sponsorship at all. The posting and the recruiter answer those. The data tells you where it is worth asking.
How to use the USCIS H-1B Employer Data Hub
The Data Hub can be queried by fiscal year, employer name, city, state, ZIP code and NAICS industry code, and you can download annual or query-specific data in Excel or CSV. Downloading is the better route, because you will want to sort and merge.
1. Start with recent years. Take the most recent full fiscal year and the partial current one. A company that sponsored heavily in 2015 and not since is a different employer now.
2. Filter by industry. The Data Hub uses NAICS sector codes. The ones most job seekers in professional roles care about, as titled by the Bureau of Labor Statistics:
| NAICS | Sector |
|---|---|
| 51 | Information |
| 52 | Finance and Insurance |
| 54 | Professional, Scientific, and Technical Services |
| 61 | Educational Services |
| 62 | Health Care and Social Assistance |
Do not filter out code 99 entirely. Some real sponsors sit there because the industry field was blank.
3. Read the right column. The Data Hub splits each employer's approvals and denials into categories taken from the I-129 petition, and each is a count of workers:
| Column | What it means, per USCIS | Why you care |
|---|---|---|
| New Employment | A beneficiary abroad with no classification, moving to a new employer in a different classification, or changing classification with the same employer | The closest signal for a student moving from F-1 status to H-1B |
| Continuation | Continuing in the same classification with no change to the job | Mostly extensions; says little about new hires |
| Change with Same Employer | A non-material change, such as a new title without a material change in duties | Ignore for list building |
| New Concurrent | Starting work with an additional employer while staying with the current one | Rarely relevant |
| Change of Employer | Moving to a new employer in the same classification | The signal if you already hold H-1B status |
| Amended | A material change to the original approved petition | Ignore for list building |
If you are a student or recent graduate, sort by New Employment approvals. If you already hold H-1B status and want to move, sort by Change of Employer. The steps for the move itself are in H1B transfer.
4. Merge the duplicate rows. Combine rows that share the employer name and the last four digits of the tax ID, so an employer with three addresses counts once.
5. Look at denials too. An employer with many approvals and a small number of denials runs a steady process. One whose denials are a large share of its filings may be worth a direct question to the recruiter about how it handles sponsorship.
6. Add the Department of Labor's data for job titles and wages. Before an employer files the petition, it has to get a Labor Condition Application certified by the Department of Labor and submit it with the Form I-129. The Office of Foreign Labor Certification publishes quarterly LCA disclosure files; the latest covers determinations from 1 October 2025 to 30 June 2026. The record layout for those files includes fields such as the employer name, job title, SOC code, worksite city, the offered wage and the prevailing wage level. That is the detail the Data Hub lacks: which titles an employer files for, where the work is, and at what wage level. A certified LCA is not a filed petition, so treat it as intent rather than outcome. The same office handles PERM labor certification for green cards, and the post on PERM processing time explains its current queue.
Cap-subject versus cap-exempt employers
Most private employers are cap-subject. The H-1B classification has an annual limit of 65,000 new statuses or visas each fiscal year, plus 20,000 more for beneficiaries with a master's degree or higher from a US institution, and cap-subject employers compete for those numbers through the registration lottery.
Some employers are outside that limit entirely. USCIS lists workers petitioned for or employed at an institution of higher education, its related or affiliated nonprofit entities, a nonprofit research organization or a government research organization as exempt from the cap. For a job seeker that means an offer from one of them does not hang on a lottery. Universities, university hospitals and research institutes are the obvious places to look, and NAICS 61 is a quick way to find the educational ones in the Data Hub. A cap-exempt employer's approvals appear in the same table as everyone else's, so check which kind of employer you are looking at before you read its New Employment count as lottery wins. Our guide to H1B cap exempt employers covers who qualifies and how to find them.
The cost picture for cap-subject employers has also changed, which may make a past sponsor more cautious now:
- The $100,000 payment. A presidential proclamation of 19 September 2025 required an additional $100,000 payment with certain new H-1B petitions. A White House fact sheet of 18 September 2026 says a new proclamation renews that requirement. USCIS's alert page records that a federal district court vacated the agency's implementing guidance on 8 June 2026, that the First Circuit denied the government's request for a stay on 24 July 2026, and that DHS will comply with the order while it considers next steps. Check that page for the current position before you rely on any summary, including this one. Our post on the H1B $100K fee covers who pays it and what it means for you.
- A proposed fee for every cap-subject petition. On 25 August 2026 DHS proposed a $103,265 fee "for all H-1B cap-subject petitions, including those eligible for the advanced degree exemption." It is a proposed rule. The comment period closed on 24 September 2026, and we found no final rule when we checked on 17 October 2026. As written it applies to cap-subject petitions, so cap-exempt employers would not pay it.
Both are costs on the employer, not on you. They are also the reason a company's sponsorship history is a weaker predictor in 2026 than it was a few years ago, and why the recruiter conversation in the last two sections matters.
The lottery numbers behind the list
A sponsor's offer gets a cap-subject candidate a registration, not a visa. USCIS publishes the registration and selection counts for each year. The last column is our arithmetic, selected divided by eligible registrations:
| Cap fiscal year | Eligible registrations | Selected | Selected share |
|---|---|---|---|
| FY2021 | 269,424 | 124,415 | 46.2% |
| FY2022 | 301,447 | 131,924 | 43.8% |
| FY2023 | 474,421 | 127,600 | 26.9% |
| FY2024 | 758,994 | 188,400 | 24.8% |
| FY2025 | 470,342 | 135,137 | 28.7% |
| FY2026 | 343,981 | 120,141 | 34.9% |
Two details in USCIS's table explain the shape. In FY2024, 408,891 eligible registrations belonged to people with more than one registration, which inflated the total. USCIS then moved to what it calls a beneficiary-centric selection process, and the multiple-registration count fell to 47,314 in FY2025 and 7,828 in FY2026. Registrations are also not people: USCIS counted approximately 442,000 unique beneficiaries in FY2025 and approximately 339,000 in FY2026.
FY2027 changed the method. DHS published the final rule "Weighted Selection Process for Registrants and Petitioners Seeking To File Cap-Subject H-1B Petitions" on 29 December 2025, effective 27 February 2026. A registration is now entered into the selection pool once for an OEWS wage level I, twice for level II, three times for level III and four times for level IV, while each unique beneficiary is still counted once. The rule's own projections put a level I registration at a 15.29 percent chance of selection, level II at 30.58 percent, level III over 45 percent and level IV over 61 percent. Those are DHS estimates made before the season, not results. Our H1B lottery post tracks registrations, odds and the wage rule.
What has been reported since: 211,600 registrations were submitted for FY2027, down 38.5 percent from 343,981, according to SHRM, citing preliminary USCIS figures. The same report says 71.5 percent of selected beneficiaries hold US advanced degrees and 17.7 percent of selected registrations fell in the lowest wage level. USCIS says it has received enough petitions to reach both the 65,000 regular cap and the 20,000 master's cap for FY2027.
The "40 percent odds" figure is not a measurement. It comes from dividing 85,000 by 211,600, which gives 40.2 percent. That division fails for three reasons. First, 85,000 is the number of cap slots, not the number of registrations selected: in every year in the table above, USCIS selected well over 85,000. Second, USCIS has not published the FY2027 selected count, so the real rate is unknown. Third, under weighted selection there is no single rate. A level I registration and a level IV registration had very different chances by DHS's own projection.
The practical lesson for your list: the wage level an employer offers now affects your chances. The LCA files show the prevailing wage level employers filed at for each title, which is one more reason to look at them. To see which level an offer falls in, the H1B wage levels guide sets out DOL's July 2026 figures for common roles in seven metros.
For context on the competition, IIE's Open Doors 2025 counted a record 294,253 students on Optional Practical Training, an increase of 21 percent from the prior year. If you are one of them, OPT to H1B covers the timeline and the cap gap.
Turning a list into applications
A spreadsheet of sponsors is not a job search. Here is how to get from one to the other.
Cut the list to roles you can win. Filter the employers down to the ones that hire for your target roles in the regions you can work in. A big sponsor in the wrong field is not a target. The job search strategy post covers how to set those roles before you start. For where international students tend to find openings, see job opportunities for international students.
Sort into three groups.
- Cap-subject employers with recent New Employment approvals in your sector. This is your main list.
- Cap-exempt employers: universities, affiliated nonprofits and research organizations. No lottery stands between an offer and a petition.
- If you are on a STEM degree, employers enrolled in E-Verify. A 24-month STEM OPT extension has to be pursued through an employer enrolled in E-Verify, and the E-Verify Employer Search lets you check. Enrollment says nothing about H-1B sponsorship, but it tells you whether a job can carry your STEM OPT time.
Go to the employer's own careers page. The Data Hub does not list open roles. Find each employer's careers site, check what is open for your roles, and read the posting for sponsorship language. A posting that says candidates must be authorized to work without sponsorship means it, even at a company with thousands of approvals. Our Tesla careers guide is one worked example, with the company's H-1B record from USCIS and DOL data.
Watch the clock. Study in the States says a student on post-completion OPT can be unemployed for a total of 90 days, and its unemployment counter page sets the STEM OPT limit at 150 days, including any days accrued on post-completion OPT. Our OPT unemployment days post lays out a 90 day search plan. That clock is why volume matters for international candidates, and why a targeted list helps you spend it on employers likely to say yes. Our post on how many applications it takes to get a job traces the published numbers behind that.
Ask early, not at the offer. When a recruiter calls, ask whether the role is open to sponsorship and whether the company has sponsored for this team. You already know from the data whether the company has done it before. The question is whether it will do it here.
Answering work authorization questions honestly
Applications commonly ask two versions of the same thing: whether you are legally authorized to work in the United States, and whether you will now or in the future require sponsorship for an employment visa. In an applicant tracking system these are usually structured fields rather than free text, which means an employer can filter on them directly. Our guide to what an ATS is covers how those knockout questions work. If you hold an H-4 EAD, the H4 EAD guide covers how to answer them in your situation.
Answer them truthfully. A student on OPT is typically authorized to work today and will need sponsorship later, and the answers should say both. The guide to answering work authorization questions has a table for 17 statuses, from citizen to OPT, H-1B, H-4 and TN. Do not answer "no" to the sponsorship question to get past a filter. The employer files the H-1B petition itself, so the answer cannot stay hidden. It surfaces at the offer, after weeks of interviews, and can cost you the offer along with the time.
If you use software to apply, the same rule applies to the software. ApplyEze reuses the work authorization and EEO answers you save, and hard facts come from your profile. When a required question is not covered by your saved answers, ApplyEze still fills it rather than stopping the run, so that answer may not come from you. Fill in your saved answers, use dry run first, and read your receipts.