Average H1B ML engineer salary is $192,000. LLM specialists at AI-native companies average $268,000+. 71% file at Level III–IV for strong lottery odds. Based on 89,400 DOL LCA filings.
The average H1B machine learning engineer salary is $192,000 per year, with a median of $185,000. These figures come from 89,400 DOL LCA filings through FY2024. ML is now the fastest-growing H1B specialty, filings grew 215% from FY2020 to FY2024, driven by the LLM era and AI-native company expansion.
The gap between AI-native employers (OpenAI, Anthropic: $285K–$295K avg) and traditional enterprise employers ($170K–$200K) is the widest of any tech specialty. For H1B strategy, targeting high-wage AI companies means both higher compensation and dramatically better lottery odds.
| Percentile | Annual Wage | DOL Wage Level | Lottery Odds |
|---|---|---|---|
| P10 (Entry) | $128,000 | Level I | 15% |
| P25 | $155,000 | Level II | 28% |
| P50 (Median) | $185,000 | Level III | 45% |
| P75 | $228,000 | Level IV | 61% |
| P90 (Top) | $285,000 | Level IV | 61% |
Source: DOL LCA filings FY2020–FY2024. Wage levels per DOL OES methodology.
California dominates at $228,000 average, 18.8% above national, due to FAANG and AI-native company concentration. Washington ($215K) is second, anchored by Microsoft, Amazon, and the booming Seattle AI startup ecosystem. Massachusetts ($212K) benefits from MIT/Harvard AI research commercialization. Texas ($188K) is growing rapidly as Austin attracts tech relocation but pays ~2% below national average.
| State | Avg Salary | LCA Filings | vs. National |
|---|---|---|---|
| California | $228,000 | 28,400 | +18.8% |
| Washington | $215,000 | 10,200 | +12.0% |
| New York | $205,000 | 8,900 | +6.8% |
| Texas | $188,000 | 7,800 | -2.1% |
| Massachusetts | $212,000 | 6,400 | +10.4% |
| Georgia | $182,000 | 4,200 | -5.2% |
| Illinois | $186,000 | 3,900 | -3.1% |
| Virginia | $195,000 | 3,600 | +1.6% |
| New Jersey | $198,000 | 3,200 | +3.1% |
| Colorado | $201,000 | 2,800 | +4.7% |
| North Carolina | $185,000 | 2,500 | -3.6% |
| Michigan | $178,000 | 2,100 | -7.3% |
The SF Bay Area commands $248,000 average, 29% above national, driven by OpenAI, Anthropic, Google DeepMind, and hundreds of AI startups. Seattle ($228K) and Boston ($222K) follow. Austin ($195K) is the fastest-growing non-coastal ML market, with Meta, Apple, and Google all establishing large engineering presences.
| City / Metro | Avg Salary | LCA Filings |
|---|---|---|
| San Francisco Bay Area, CA | $248,000 | 18,600 |
| Seattle, WA | $228,000 | 8,900 |
| New York City, NY | $218,000 | 7,200 |
| Los Angeles, CA | $212,000 | 5,400 |
| Boston, MA | $222,000 | 4,800 |
| Austin, TX | $195,000 | 3,900 |
| Chicago, IL | $188,000 | 3,200 |
| Washington DC | $198,000 | 2,900 |
| Denver, CO | $204,000 | 2,600 |
| Research Triangle, NC | $188,000 | 2,200 |
71% of H1B ML engineer filings are at Level III or IV. AI-native companies (OpenAI, Anthropic, Nvidia) file 64–78% at Level IV, making them among the best lottery-odds employers available. Entry-level ML positions (Level I: 15% odds) are the worst possible H1B strategy; they're both lower paid and far less likely to be selected. If you have 2+ years of ML experience, insist on Level III minimum.
✓ Strategy tip: An ML engineer at a Level IV AI company has a 61% lottery selection rate, vs. 15% at a Level I consulting shop filing the same role. The employer and wage level you choose determines your H1B odds more than any other factor.
AI-native companies have created a premium wage tier that sits 50–80% above traditional enterprise ML pay. Anthropic ($295K avg) and OpenAI ($285K) lead the market, with 72–78% Level IV filing rates. Nvidia ($265K) reflects the GPU hardware-to-AI convergence premium. FAANG companies average $228K–$255K with 48–62% Level IV rates.
| Employer | LCA Filings | Avg Wage | % Level IV |
|---|---|---|---|
| Google / Alphabet | 6,800 | $248,000 | 58% |
| Meta | 5,400 | $255,000 | 62% |
| Amazon / AWS | 7,200 | $228,000 | 48% |
| Microsoft | 6,100 | $232,000 | 51% |
| Apple | 4,200 | $242,000 | 55% |
| OpenAI | 1,800 | $285,000 | 72% |
| Anthropic | 920 | $295,000 | 78% |
| Nvidia | 3,400 | $265,000 | 64% |
| Salesforce | 2,800 | $218,000 | 44% |
| Adobe | 2,100 | $212,000 | 41% |
No H1B specialty has grown faster than ML engineering. Filings jumped from 28,400 in FY2020 to 89,400 in FY2024, a 215% increase. The ChatGPT/GPT-4 inflection point in FY2023 drove the sharpest single-year increase (+25.6%). Average wages rose 29.7% over the same period. The market shows no signs of cooling with frontier model training, AI infrastructure, and enterprise AI deployment all requiring ML expertise at scale.
| Fiscal Year | LCA Filings | Avg Wage | YoY Change |
|---|---|---|---|
| FY2020 | 28,400 | $148,000 | , |
| FY2021 | 42,800 | $162,000 | +9.5% |
| FY2022 | 58,900 | $172,000 | +6.2% |
| FY2023 | 74,200 | $182,000 | +5.8% |
| FY2024 | 89,400 | $192,000 | +5.5% |
LLM/foundation model engineers now earn a $70K–$80K premium over classical ML roles. This reflects the concentration of LLM talent at high-wage employers (OpenAI, Anthropic, Google DeepMind) and the genuine scarcity of engineers with pre-training, RLHF, and model alignment experience.
MLOps and ML platform engineers, the "infrastructure for AI" layer, are increasingly critical and command $200K–$220K. Applied ML engineers at product companies range $185K–$215K depending on seniority.
| Specialization | Avg H1B Wage | % Level IV | Market Growth |
|---|---|---|---|
| Large Language Models (LLM) | $268,000 | 70% | Explosive |
| Computer Vision | $228,000 | 58% | High |
| NLP / Text Processing | $222,000 | 55% | High |
| Reinforcement Learning | $245,000 | 64% | High |
| MLOps / ML Platform | $208,000 | 48% | High |
| Recommendation Systems | $215,000 | 50% | Moderate |
| Applied ML (Product) | $198,000 | 42% | Moderate |
| Research Scientist | $252,000 | 65% | High |
ML engineers are in the strongest negotiating position of any H1B category right now. Demand for LLM expertise dramatically outpaces supply. Use that leverage. The LCA database shows exactly what each employer paid ML engineers in your target metro last year, that's your floor, not your ceiling.
Target AI-native over enterprise
OpenAI, Anthropic, Nvidia, and Cohere pay $70K–$100K more than enterprise employers for equivalent ML roles and file at higher wage levels. Better pay AND better lottery odds.
Specialize in LLM/foundation models
If you can credibly claim pre-training, RLHF, or alignment experience, you qualify for the $250K+ premium tier. Even post-training/fine-tuning experience at scale commands a premium.
Use competing offers aggressively
ML is one of few tech specialties where you can realistically hold multiple offers simultaneously. A written offer from an AI company is the strongest possible leverage with any employer.
Negotiate equity separately from LCA wage
The LCA only covers base salary. Equity (RSUs at public companies, options at startups) can double or triple total compensation. These are negotiated entirely independently of H1B requirements.
Browse ML and AI positions at companies with strong H1B sponsorship and Level IV wages.
H1BVisaJobs.com Research Team
Analysis based on 89,400 DOL LCA filings for ML engineering roles (FY2020–FY2024), cross-referenced with BLS OES Computer Occupations data and AI company public filings. Last updated: May 2026.