Hiring in New York vs San Francisco: What Actually Differs
San Francisco pays more and has deeper senior AI talent. New York has broader industry spread and less competition per candidate. Both are expensive.
If you are choosing between opening a role in New York or San Francisco, the short answer is that San Francisco still pays more and still holds the deepest concentration of senior AI and infrastructure engineers, while New York gives you a broader industry base and less competition for each candidate. Both are expensive. Neither is the budget option. What should decide it is not the salary band but how many other companies are already talking to the person you want, and how quickly you can reach a decision.
Before Pickr I ran a recruitment agency. The failure I saw most often in US searches was a team that picked a city off a two-year-old salary benchmark and then lost every candidate at offer stage. The benchmark was rarely the problem.
Which city pays more for engineers?
San Francisco, still, but by less than most budgets assume. For a comparable senior software engineering role, San Francisco compensation commonly runs about 10% to 20% above New York. For senior AI, ML infrastructure and platform roles the gap is wider again, and at the very top of that market total packages have detached from anything a normal budget recognises — those outliers are not a benchmark you can plan against.
Two things compressed the gap. New York pay rose quickly over the past few years, pushed by finance and quant firms that never competed on equity and simply paid cash. And candidates now discount the equity component of a Bay Area offer far more heavily than they did in 2021, so the headline number and the number the candidate actually values sit further apart than your spreadsheet says.
Practical version: benchmark against the city, not against a national US median. A national figure will leave you light in both.
Where is competition per candidate worse?
This matters more than the salary band, and it favours New York.
San Francisco has more senior AI and infrastructure people, and a much higher density of employers chasing exactly that profile. My planning assumption for a strong senior infrastructure engineer in the Bay Area is 3 or 4 live conversations already running. In New York it is more often 1 or 2, because demand is spread across finance, media, health tech and adtech instead of concentrated on one archetype.
That changes the arithmetic of a search. In San Francisco you are rarely competing on whether the candidate likes your role — you are competing on whether you reached a decision before three other companies did. In New York you have more room to run a proper process and less excuse for being slow, because the finance side of that market moves quickly and does not wait for your panel to reconvene.
| San Francisco | New York | |
|---|---|---|
| Senior AI / infra depth | Deepest anywhere | Good, growing |
| Industry spread | Narrow, tech-concentrated | Broad — finance, media, health tech |
| Competition per candidate | Assume 3 to 4 live processes | Assume 1 to 2 |
| Equity expectation | High, central to offers | Lower, cash-weighted |
| Counter-offer style | Equity refresh, often accepted | Cash and bonus, decided fast |
| Sector-downturn exposure | High | Diversified |
Treat that table as planning assumptions from running searches, not as survey data. The direction is reliable. The exact numbers are not, and they move with the funding cycle.
Does the counter-offer culture differ?
Yes, and it catches European hiring teams out. Both cities counter aggressively enough that you should plan for one on every senior offer rather than treat it as bad luck.
The difference is shape. San Francisco counters arrive as accelerated equity refreshes and retention grants, and they land well, because the candidate is being asked to compare one unvested pile against another and the incumbent holds the information advantage. New York counters are more often straight cash — a bonus pulled forward, a title — and get resolved in days rather than weeks.
The fix is the same in both markets and it is not clever: raise the counter-offer conversation 2 weeks before the offer, not on the day of resignation. If you have never asked a candidate what their employer will do when they resign, you have not run the process.
How do notice periods change the timeline?
2 weeks, both cities. This is the largest structural difference from European hiring and it changes how you plan.
A US start date is realistically 4 to 6 weeks from acceptance. In Germany or Austria the same hire is often 3 months out. That sounds like a pure advantage until you notice the other side of it: a candidate who can start in 2 weeks can also accept a competing offer in 2 weeks. There is no long exit window in which a slow employer catches up.
If your process runs 30 to 45 days end to end, which is where most in-house processes land, you are too slow for either city. Under 21 days from first contact to offer is the bar. That usually means fewer stages rather than faster stages, and a real decision at the end of each one. I have written separately about where the time actually goes in a slow hiring process; it is almost never the interviews themselves.
Has remote hiring made the city question obsolete?
No. It changed the question.
Compensation bands, onsite expectations and the density of senior specialists are all still local. What stopped being local is your competition. A remote-eligible New York role competes with every well-funded remote employer in the Eastern time zone and a good number in the Central. The same role out of San Francisco competes across the Pacific band, including companies willing to pay Bay Area rates to someone living in Denver.
So the question stopped being where do we hire and became who else is talking to this person. Most hiring teams have no way to answer that. The available signal is behavioural: how fast candidates go quiet, how often you lose people between second stage and offer, and how many rejections arrive with a competing offer attached.
Return-to-office posture is now a real filter too. A five-day onsite mandate cuts your Bay Area pool harder than the same mandate in New York, because more Bay Area engineers moved out during 2021 and 2022 and plenty of them stayed out. Decide that before you write the brief, not after a candidate withdraws over it.
Which market should you open the role in?
Open in San Francisco if the role genuinely needs someone who has done the thing at scale — trained large models, run inference infrastructure, built a platform serving hundreds of engineers. That depth is there and it is not reliably anywhere else. Accept the premium, move inside 3 weeks, and expect to lose some candidates anyway.
Open in New York if the role is a strong generalist, if the budget is finite, or if you want a market that does not empty out when one sector corrects. More names per search, more room for a real process, and candidates who are less conditioned to treat every offer as an opening bid.
Open it remote only if you can run a structured process across time zones. Remote hiring exposes a weak process faster than onsite hiring does, and most teams find that out on a candidate they wanted.
How Pickr helps when you are hiring in both cities
Pickr is an AI-native recruiting platform for recruitment agencies and in-house teams. It scores candidates on evidence of skills rather than keyword matches, including adjacent and transferable ones, which is the specific problem a two-market search creates: a New York fintech engineer and a Bay Area infrastructure engineer often hold the same capability and almost never describe it in the same words. What happened to the people you actually hired feeds back into how the next candidates are evaluated, so a team hiring into both cities gradually learns which signals travel and which are local noise.
Interviews are transcribed and scorecards arrive pre-filled with evidence mapped to each criterion, so the interviewer edits a draft instead of facing an empty form days later. When the whole timeline is 3 weeks, that is not a convenience. It is the difference between deciding and drifting. Interviewer and hiring-manager seats are free, so pulling in the two extra people who need to weigh in costs nothing.
Two honest limits. That feedback loop needs outcomes to learn from, so on a first search in a new city it is doing less for you than it will after a dozen placements. And candidate data is hosted in Frankfurt, Germany, with Pickr built in Austria — which matters if you are a European company hiring into the US and your own compliance obligations follow the data rather than the candidate, and is a neutral fact if you are a US-only team that does not care where the servers sit.
New York or San Francisco: the short answer
San Francisco buys you depth at the senior AI and infrastructure end, charges roughly 10% to 20% more for it, and hands you a process where several other companies are already in the conversation. New York buys you breadth, resilience across sectors and noticeably less competition per candidate, at a somewhat lower price. Neither is cheap. Neither forgives a slow process.
The decision that moves your hit rate is not the city. It is whether you can get from first contact to offer inside 3 weeks with a real decision at every stage. Teams that manage that do well in both markets. Teams that cannot fail in both, then blame the market.
Frequently Asked Questions
Is San Francisco or New York more expensive for engineering hires?
San Francisco is still more expensive at the senior end, commonly around 10% to 20% above New York for a comparable software engineering role, and the gap is wider again for senior AI and infrastructure people. At junior and mid levels the two markets are close enough that compensation should not decide where you open the role. New York pay has risen quickly in recent years, so the historical gap is narrower than most budget assumptions still allow for.
Which market has more engineering candidates available?
New York has the larger overall technology workforce because it spans finance, media, health tech, adtech and retail, while San Francisco has the deeper concentration of senior AI, infrastructure and platform engineers relative to its size. If you need someone who has run inference infrastructure in production, San Francisco has more of them. If you need a strong generalist backend engineer, New York gives you more names and less competition per name.
How do counter-offers differ between New York and San Francisco?
Counter-offers are common in both markets but take different shapes. San Francisco counters usually arrive as equity refreshes or retention grants and are frequently accepted, so a signed offer is less final than it looks. New York counters are more often cash and bonus driven, particularly in finance, and get resolved in days. Plan for a counter on every senior offer rather than treating it as bad luck, and raise the subject before the offer goes out.
What notice periods should I expect in US tech hiring?
Both New York and San Francisco run on 2 weeks notice as standard, far shorter than the 1 to 3 months common across most of Europe. That makes a US start date realistically 4 to 6 weeks from offer acceptance rather than 3 months out. The practical consequence is that speed of decision matters more in the US, because candidates are rarely locked into a long exit window that gives a slower employer time to catch up.
Has remote work made the New York versus San Francisco choice irrelevant?
Remote hiring did not make the choice irrelevant, but it did change the question. It moved the decision from where you hire to who you are competing with, because a remote role in either city competes with every well-funded remote employer in the same time zone band. Location still sets compensation benchmarks, onsite expectations and the density of senior specialists, but the candidate pool is no longer bounded by commuting distance from either city.
Free recruiting audit · 2 minutes
Find out what your hiring process is actually costing you.
Answer eight questions, or connect your current system read-only, and get a report on where your funnel loses candidates and which changes are worth making. No signup, no API key stored, data stays in the EU.
Written by Andreas Amann
Founder of Pickr. Former operator at startups in Berlin and Silicon Valley, where he helped scale companies from 40 to 200+ people. Built Pickr after years of using every major ATS as a recruitment agency owner at ScalingPPL.