Blog
Emerging H-1B Sponsors to Watch: Tesla, Nvidia & the New Wave
September 4, 2026
A new class of product and AI companies is filing thousands of H-1B labor conditions a year while the old IT-services order recedes. Here's who to watch, and how to read their numbers.
For two decades, the list of top H-1B sponsors read like a directory of IT outsourcing firms. The names at the top filed tens of thousands of labor condition applications a year, mostly for contract programmers and systems analysts. That order is shifting. A new group of product-first and AI-first companies is now filing at scale — and in the most recent partial year, the single largest sponsor is a retailer-turned-cloud-giant, not an outsourcer.
This article looks at the challengers: who they are, how much they file, and — just as important — how to read the numbers without drawing the wrong conclusion. If you are weighing an offer, tracking the market, or just curious where hiring is heading, these are the sponsors worth watching.
The incumbents are still huge — but no longer the whole story
The traditional leaders still dominate cumulative filing counts. Firms like Infosys, Tata Consultancy Services, and Cognizant built their US presence on placing large numbers of workers at client sites, and their all-years totals reflect that. We compared that outsourcing model against Big Tech in detail in Outsourcing Giants vs Big Tech.
But the top of the leaderboard is no longer static. In fiscal year 2026 — a partial year, with only about two quarters reported so far — the top sponsor is Amazon, a product and cloud company, not a staffing intermediary. The chart below shows the current dataset-wide ranking of sponsors by filing volume.

The new wave: five sponsors to watch
The interesting movement is one tier down, among companies whose H-1B footprint has grown alongside their products. These are not staffing firms; they hire engineers, scientists, and researchers to build things in-house. Here are five, with their all-years labor condition application (LCA) filing totals and certification rates from the disclosure data:
| Sponsor | All-years LCA filings | Certification rate |
|---|---|---|
| Tesla | 10,903 | 56.9% |
| Nvidia | 7,273 | 65.4% |
| ByteDance | 3,468 | 92.3% |
| ServiceNow | 3,566 | 67.4% |
| Palo Alto Networks | 2,722 | 56.8% |
A few things stand out. Tesla leads this group on raw volume — nearly 11,000 filings across all years — reflecting a hardware, software, and manufacturing workforce that spans far more than a typical software shop. Nvidia, the company at the center of the AI compute boom, sits second with more than 7,000 filings, and its rise tracks the broader surge in demand for chip and AI talent.
ByteDance, ServiceNow, and Palo Alto Networks each file in the low thousands. That is small next to the outsourcing incumbents, but it is the shape of a product company staffing engineering and research roles directly rather than through a contractor pipeline. Watch these counts over the next few years: this is the tier where the fastest relative growth tends to show up.
Read the certification rate carefully
The certification rate column above is where most casual readers go wrong, so it is worth slowing down.
An LCA is not a visa. It is a filing an employer submits to the Department of Labor attesting to the wage and working conditions for a role. Certification of that filing is not the same as a USCIS visa approval — the two are separate steps run by separate agencies. A high certification rate does not mean a company "wins" more visas, and a lower one does not mean workers were turned away at the border. We unpack this distinction in Who Sponsors the Most H-1Bs — and Who Has the Cleanest Record.
More specifically: the mid-range certification rates you see for Tesla (56.9%) and Palo Alto Networks (56.8%) are largely a certified-withdrawn pattern, not rejections. Employers routinely file more positions than they ultimately use, or file duplicate and precautionary applications and then withdraw them. A withdrawn filing is not a denied one. So a company at 57% is not being blocked from hiring at nearly the rate a naïve reading suggests — it is filing broadly and pruning later.
ByteDance's 92.3% rate sits at the other end. A very high rate typically reflects a tighter, more targeted filing practice rather than any special favor. In other words, the number tells you something about how a company files, not how "good" a sponsor it is.

Why AI and product companies are rising
The through-line connecting Tesla, Nvidia, ByteDance, ServiceNow, and Palo Alto Networks is that they compete for a specialized, in-house technical workforce — increasingly weighted toward AI, data, and security roles. Where the old model was about supplying large volumes of interchangeable contract labor, the new wave is about hiring scarce specialists directly.
That maps onto occupation-level trends in the data. Roles such as Data Scientists and Software Developers have become central to sponsorship in exactly the segment these companies occupy — a shift we traced in The Rise of the Data Scientist. The chart below shows the occupations drawing the most H-1B filings across the dataset, which helps explain where the new wave is concentrating its hires.
How to watch these sponsors yourself
If you want to track the challengers rather than take a snapshot at face value, keep a few habits:
- Compare filing volume year over year, not just cumulative totals. All-years counts favor companies that have filed for a long time. A newer, fast-growing sponsor can look small in the cumulative figure while climbing quickly in recent years.
- Treat FY2026 as provisional. It is a partial year — roughly two quarters reported — so any ranking that leans on it will move as more data lands.
- Separate the wage you see from the wage you'd take home. The "offered wage" on an LCA is a gross prevailing-wage figure, not a paycheck. It is useful for benchmarking a role, not for estimating net pay.
- Don't read certification rate as approval rate. As above, mid-range rates here reflect certified-then-withdrawn filings, not USCIS decisions.
You can dig into any single employer's own filing history, occupations, and locations on its profile page, and use the sponsor finder to line up companies against a specific job title and state.
The takeaway
The H-1B leaderboard is being reshaped from the top down. The outsourcing giants remain enormous by cumulative volume, but the growth story now belongs to product and AI companies — Tesla and Nvidia at the front, with ByteDance, ServiceNow, and Palo Alto Networks filling in behind them. FY2026's provisional top spot going to Amazon rather than an IT-services firm is the clearest signal yet that the new wave is real.
Just remember what the numbers are and are not. Filing counts measure activity, not headcount. Certification rates measure how a company files, not whether its workers get visas. Read them that way, and the shape of the next few years comes into focus.
- new H-1B sponsors
- Tesla
- Nvidia
- AI hiring
- certification rate