career · career
PM salary by city: SF, NYC, Seattle, Austin, London, and remote (2026)
The city you pick in 2026 is also an AI-exposure decision. SF and Seattle pay more in absolute terms, but the gap between those markets and everywhere else is partly explained by role mix: the AI PM titles that carry a 15-20% premium concentrate there, and the senior AI PM total comp range ($320-520K) exists almost exclusively in those two metros. Everything below is sourced from Levels.fyi EOY 2025 data, Glassdoor aggregates, and company-specific reports. Numbers are 2026 medians unless noted.
Base and total comp by city
| City | Level | Base | Total comp |
|---|---|---|---|
| SF Bay Area | APM / L4 | $155K | $274K |
| SF Bay Area | Senior PM / L5 | $210K | $371K |
| SF Bay Area | Staff PM | $242K | $430K+ |
| NYC | APM | $141K | $195K |
| NYC | Senior PM | $184K | $295K |
| NYC | Staff PM | $243K | $380K |
| Seattle | APM | $135K | $185K |
| Seattle | Senior PM | $168K | $310K |
| Seattle | Staff PM | $212K | $390K |
| Austin | APM | $114K | $148K |
| Austin | Senior PM | $143K | $190K |
| Austin | Staff PM | $183K | $240K |
| Remote (US) | APM | $130K | $165K |
| Remote (US) | Senior PM | $159K | $188K |
| London | Mid-level | £83K | £106K (~$134K) |
| London | Senior | £120K | £145K (~$183K) |
| London | Top decile | £155K | £181K (~$228K) |
Google APM L4 median total comp is $274K. Big Tech Senior PM base across Meta, Google, Amazon, Apple, Microsoft, Stripe, and Databricks runs $215K-$260K. Meta London is the London outlier at approximately £301K total comp for senior levels: roughly what a mid-level PM earns at US FAANG.
Purchasing-power reality check
SF pays approximately 35% more in base than Austin. But SF cost of living is roughly 79% higher. On a Senior PM base of $210K (SF) vs $143K (Austin), the SF premium largely evaporates once you adjust for housing. Austin and remote-from-Austin are viable for applied-AI feature work. They are not viable if your target role involves model ownership, eval design, or agent architecture: those roles are disproportionately SF and Seattle, and they are the roles that actually command the top of the comp distribution.
Seattle is the most underrated market in this comparison. Cloud equity at Amazon, Microsoft, and the growing Seattle AI cluster (AI2, Cohere’s US office) pushes Staff PM total comp to $390K or higher, with a CoL that is roughly 30% below SF. SF-anchored remote roles (companies that set pay to SF bands regardless of your location) pay about 9% more than median remote, which closes some of the Seattle gap if you can land one.
Remote: location-agnostic vs location-tiered
Most public companies apply geographic tiers. Amazon, Meta, Google, Apple, and Microsoft require three to five days per week in office as of 2026. Remote PM roles at these firms are rare exceptions. If you are negotiating remote at a big tech company, the leverage is almost entirely on your side at Staff and above, not at APM or Senior PM.
Three pay models determine what you actually earn as a remote PM:
- Flat national band. GitLab, Automattic, and Zapier pay the same regardless of US metro. A PM in Austin earns the same as one in SF on the same band.
- CoL-adjusted. Google, Meta, Microsoft, and Amazon benchmark to a reference city (usually SF or Seattle) and discount by CoL tier. Moving from SF to Austin on this model means a pay cut.
- Hub-tier. Some Series B and C startups in 2026 pay an above-remote-national rate for hybrid roles in secondary markets (Austin, Denver, Chicago), creating a middle tier between “big tech office” and “fully remote anywhere.”
Ask before the offer stage: “What pay model does your company use for remote roles, and which tier does my location fall into?”
London vs US: the real gap
A London Senior PM at a top non-Meta firm earns roughly what a mid-level PM earns at US FAANG in absolute dollars: approximately $183K total comp vs $295K in NYC or $371K in SF. The purchasing-power gap narrows somewhat because London CoL, while high, sits below SF. It does not close.
For European AI company roles (DeepMind, Wayve, Stability, Mistral), London is the right market, and those companies offer equity upside that changes the math at senior levels. But if you are choosing between a US remote role and a London offer, the US remote role earns 2.2-2.7x more in absolute terms. The London vs US gap is the comparison that almost no competing salary guide includes: the numbers above are the ones to use when evaluating a cross-Atlantic move.
The PM vs engineering manager trajectory
Levels.fyi EOY 2025: PM comp grew +4.55% year over year while SWE Manager comp grew +9.64%. PMs are falling behind the management track in absolute comp growth. The countervailing argument is that AI PM roles at frontier labs are compressing that gap at senior levels, but only for a narrow set of roles at a narrow set of companies. If your goal is total comp maximization over five years, the city and role-type choice matters more than the negotiation tactics on any single offer.
AI PM premium by city
The $25-34K AI PM premium at mid-level holds in SF, Seattle, and NYC. In Austin, the premium exists but the absolute numbers are lower because the base market is lower. Remote AI PM roles tend to be applied-feature work rather than core-model roles; the premium there is closer to 10-15% rather than 20%.
The $320-520K total comp range for senior AI PM is not remote-compatible at most organizations. Those roles require proximity to model teams, which means SF or Seattle. SF AI PM median total comp is materially higher than Austin AI PM at equivalent seniority, not because of a comp policy difference but because the role types that command that comp do not exist in the Austin market at the same density.
For negotiation tactics once you have an offer, see negotiate equity, not base and AI PM salary in 2026. For comp broken out by level rather than city, see PM salary by level.