Machine Learning Engineer salary in San Francisco & total compensation (2026)

Why ML Engineers sit a band above generalist AI engineers

The Machine Learning Engineer still owns the model-training half of the stack — pipelines, feature stores, experiments, and evaluation infrastructure for production models — rather than only consuming foundation-model APIs. That concentration at frontier labs and large tech is why the Levels.fyi ML Engineer median ($272k) sits materially above the broad-market AI Engineer title. At the top, Anthropic's Business Insider-reported technical-staff bands reach up to $1.38M base under the "Member of Technical Staff" umbrella, with reinforcement-learning researchers specifically at $112k–$500k base.

Machine Learning Engineer pay in San Francisco

San Francisco is the price-setting anchor for AI comp: Levels.fyi's Wrapped 2025 put median SWE total comp at $278k in the Bay Area versus $193k in New York — a ~44% gap that has not closed. The metro holds about 35% of US AI engineers, because frontier-lab HQs and NVIDIA all anchor here. The trade-off is the steepest cost of living in the country (~20% above the US average), so COL-adjust before comparing a Bay number to anywhere else.

Machine Learning Engineer pay by seniority

Level is the single biggest driver of a Machine Learning Engineer offer — a one-level move can change total comp by tens of thousands a year. The bands below are computed from posted pay-transparency ranges in our index:

• Mid (~$353k) — well-scoped work with guidance, ramping to independent delivery.
• Senior (~$225k) — owns features and systems end to end; where most experienced hires land.
• Staff (~$275k) — drives architecture and tradeoffs across teams.
• Principal (~$430k) — sets technical direction; comp here is individualised and equity-dominated.

The reinforcement-learning and infra premium

Within ML engineering the pay split is driven by two things: RL / frontier-model-training experience, and GPU-pipeline ownership at hyperscale. Anthropic prices RL researchers into the MTS band precisely because that skill is scarce. At Databricks the "Staff Machine Learning Engineer" is explicitly a platform-pipeline owner rather than a model researcher — a reminder that the same title means different work (and different bands) at a lab versus an enterprise-data company.

AI-native vs traditional pay for a Machine Learning Engineer

The same title is priced differently depending on whether the employer is AI-native. Comparing posted bands for Machine Learning Engineer roles in our index at AI-native companies against everyone else:

That 12% premium is the single best reason to position yourself as AI-native rather than "a software engineer who also uses AI." The framing alone changes which band a recruiter benchmarks you against.

How the hiring bar prices the role

Frontier-lab ML loops test "ML-native coding fluency" — implementing attention or a training loop in PyTorch/NumPy and diagnosing a broken net — rather than LeetCode grinding. That bar is why the role's floor is high: labs recruit on demonstrable model engineering, and acceptance rates sit under 1%. Mainstream AI startups lean more on coding + system design + applied-ML case studies, which is part of why their bands trail the labs.

What an ML Engineer actually does in 2026

The ML Engineer still owns the model half of the stack: data preprocessing, feature engineering, training and hyperparameter tuning, and the evaluation infrastructure that gates production models. The work is concrete — at OpenAI, ML Engineers on Integrity teams monitor and maintain deployed models so they keep delivering value in production. That model-training centre of gravity is what separates the role from an AI Engineer who mostly integrates foundation-model APIs, and it is why the median lands a band higher.

Career ladder: from software or stats into ML

ML engineers typically come from software engineering or an applied-statistics background and progress Senior (L4) → Staff (L5) → Principal (L6). At Anthropic the ML-research track sits inside the Member of Technical Staff ladder, so a strong ML engineer with research instinct can move toward research engineering rather than only up the platform track. The fork in the road is whether you deepen into modelling/RL (toward research) or into pipelines and GPU infra (toward platform/staff) — both pay well, but they are recruited differently.

Which companies hire Machine Learning Engineers in San Francisco

The most active AI-native hirers for this role in our index right now — a company scaling a function usually means clearer levelling and more room to negotiate. Each links through to its open roles and comp range:

Negotiating an ML Engineer offer

The published lever ranking is consistent across 500+ negotiated AI-lab offers: level is the highest-impact lever — an L4→L5 jump at a frontier lab moves total comp by $150k–$300k a year — and a written competing offer is "the single most reliable" way to move a number. Base salary is the hardest component to shift: both OpenAI and Anthropic run strict bands, so anchor on level and equity, not base. If your experience is RL or training-infrastructure heavy, benchmark against the research-engineering band rather than generic ML engineering — the two overlap on comp but are recruited differently.

How we calculated these numbers

Every salary figure on this page is computed from pay-transparency bands posted in live job listings in our index — 38 Machine Learning Engineer postings in San Francisco currently disclose a band, out of 535 live matching roles. Where a listing posts a range we take the midpoint, and we refresh weekly. These are calibrated ranges, not offers.

Sources

• Levels.fyi — Machine Learning Engineer title
• Business Insider — Anthropic salaries revealed (2026)
• How to get hired at OpenAI, Anthropic & DeepMind in 2026