Software Engineer Jobs in San Francisco: 2026 Salary, Top

San Francisco is still the highest-paying market for software engineers in the world, but the hiring channel has split in two. The published-jobs lane is overcrowded — around a thousand applicants land on a typical Series A senior posting in the first week. The matched-and-sourced lane runs quieter and converts faster. Most senior engineers spend three months in the wrong lane before they figure this out. We built Standout because the application funnel stopped working for this market specifically. Here's the 2026 snapshot of software engineering jobs in San Francisco — what they pay, who's hiring, and how to actually get one.

For the AI engineering subset of this same market, see AI engineer jobs in San Francisco. For React-focused frontend hiring across the same companies, see React engineers 2026.

TL;DR — Software Engineer Jobs San Francisco, May 2026

The application math has not gotten better. The matching math has gotten dramatically better. That's the entire story.

What's actually happening in SF software engineer hiring

The market in May 2026 is two distinct economies running at the same time. Treat them as one and the search will take six months.

The published-postings economy is flooded. A senior backend role at a Series B SF company pulls 600-1,200 applications in the first seven days, the vast majority of which never reach a recruiter — 88% of companies use AI screening at the first pass and around 75% of resumes auto-reject before human review. This is the lane most candidates default to and it converts at 2-3% in current conditions for senior roles.

The sourced economy runs underneath. Hiring managers at the companies worth working at have lost trust in their inbound pipelines. They'd rather wait for a referral, a recruiter-sourced intro, or a matched candidate from a talent agent. Across the matches we've run on Standout, the conversion from intro to first call sits in the 60-80% range, and the conversion from first call to offer is materially better than from cold applications. The candidates who switch lanes go from "two phone screens after a hundred applications" to "five founder intros in a month."

Hot take: if you've sent more than fifty cold applications to SF software engineering roles in the last ninety days and have zero offers, the problem is the lane, not your resume. Stop sending applications and switch lanes. Period.

Top hiring companies in San Francisco — tiered

Hiring volume is concentrated. About 70% of the offers we see senior candidates run come from a small set of companies. Skip the rest until you're done with these.

Tier 1: AI-native juggernauts and frontier labs

Anthropic (2,500 staff and growing fast), OpenAI, Google DeepMind, Meta Superintelligence Labs, xAI, Cursor, Perplexity. These run the highest comp ceilings in the market. Total comp packages of $400K-$1M+ are normal for senior to staff engineers, and they hire generalist software engineers, not just ML specialists. Most of their hiring at senior+ levels happens through direct sourcing or executive networks. The job board is a small fraction of actual hiring.

Tier 2: High-velocity scale-ups

Stripe, Databricks, Anduril, Ramp, Notion, Figma, Vercel, Plaid, Mercury, Linear, Retool, Sierra, Glean, Harvey, Hippocratic. These companies hire continuously, pay aggressively (mid-level $300K-$450K, senior $400K-$650K), and run more structured interview processes. They post most of their roles publicly, but warm intros materially shorten the timeline. The funnel is competitive but predictable.

Tier 3: Big tech in SF

Google, Meta, Apple, Microsoft, Amazon, Salesforce, Adobe. Comp is steady at $250K-$450K for mid-to-senior. The hiring funnel is structured and reliable, applications work, but cycle times are 8-14 weeks. The equity upside is muted compared to the scale-up tier. Best fit for engineers who optimize for stability and structured progression.

Skip these for now

Pre-product-market-fit startups (under 15 people) with sub-market comp. Generalist enterprise companies hiring "software engineers" at $150K base in 2026 — they are 2-3 standard deviations below the market and the equity won't make it up. Read the comp band before investing time in any application.

Salary by experience — what SF software engineers actually make

These ranges are total compensation (base + RSUs at current preferred or public price + bonus) for full-time SF roles. The numbers come from Levels.fyi, public Indeed data, and the offers we see candidates run.

Levels.fyi puts the SF software engineer median at $272,550, with the 75th percentile at $375K and the 90th percentile at $500K. Indeed reports a base salary average of $176,198 across SF software engineers, which tracks for non-FAANG roles but understates the FAANG and AI-native scale-up bands considerably.

Hot take: if you have 5+ years of experience and your most recent SF offer was under $300K total comp, you're being underpaid by current SF market rates. The market for senior engineers is not soft right now. Counter or walk.

In-demand skills — what hiring managers actually filter for

Across the senior software engineering matches we've run, the responsibilities cluster tighter than the public job descriptions suggest. The market filters for these:

1. 1Distributed systems fluency. Most SF roles at scale-ups deal with horizontal scale issues from week one. Hiring teams want to see that you've owned a service handling >10K RPS or built a system that survived a real incident. Generic "I worked on a microservices architecture" loses to "I designed the sharding layer that took us from 5K to 50K writes per second."
2. 2TypeScript / Python depth. Python and JavaScript continue to lead language usage on GitHub, with TypeScript climbing to #3 globally. At the SF scale-ups, TS on the application layer and Python on the data/ML layer is the dominant stack. Go for infrastructure. Rust is rising fast but still niche.
3. 3Cloud infra (AWS or GCP). Most SF tech is on AWS, with a meaningful tail on GCP at the AI-native scale-ups and Anthropic. Native AWS familiarity is now table stakes for senior backend roles.
4. 4LLM-application work, even at non-AI companies. This shifted in late 2025. Even a generalist backend role at a fintech now expects you to have shipped or co-shipped at least one LLM-powered feature. Engineers without any LLM exposure are getting filtered at scale-up tier.
5. 5Code quality at scale. Hiring teams care about engineers who write maintainable, well-tested code under throughput pressure. Coding interviews now lean more design-and-tradeoff than algorithm-puzzle. Practice the design loop.
6. 6Public artifacts. A maintained GitHub, a technical blog, an open-source contribution. The candidates we represent who have public artifacts run 2-3x the response rates on intros. This is structural — hiring managers want to see how you think before they spend an hour with you.

What matters less than the market thinks: leetcode bins, big-tech pedigree alone, certifications. The bar is shipping evidence, not credentials.

Hot take: the engineers at the strongest SF startups have the smallest LinkedIn followings and the most active GitHub repos. Personal brand on LinkedIn is mostly a tax in this market. A maintained repo with non-trivial code beats 10K followers every time. Don't optimize for the wrong asset. See how to get recruiters to come to you for the full inbound playbook.

How to actually get matched in San Francisco

The application channel is the worst it has been in two decades for senior tech roles. The matching channel is the best it has ever been. The right strategy is to invert the funnel — let companies come to you. For the broader cross-channel SF playbook, see how to find startup jobs in San Francisco.

What works in 2026

1. 1Build a profile on a talent agent once. Standout matches you with US tech companies hiring across all roles — engineering, product, design, data, ML/AI, ops. We intro you directly to the founder or hiring manager when you say yes to a match. First matches typically arrive within a few hours of profile completion. Free for candidates.
2. 2Pick three or four dream companies and write to people who work there. Not recruiters. Engineers. About something specific they shipped. Four-sentence email, specific reason you're useful to their team. This converts at 5-10% but the ones who reply move fast.
3. 3Get on the right Slack communities and in-person events for your discipline. Warm intros convert at 4-5x the cold-application rate. Most senior candidates underbuild their network before they need it.

What to stop doing

Stop using auto-apply tools. They flag as spam to ATS scoring models and signal "spray and pray" to the recruiters you want to reach. Stop the public LinkedIn Open to Work badge — recruiters at the companies worth working at read it as an anti-signal. Stop tailoring resumes for ATS keyword matches. The hours invested don't move the response rate above 3% in this market. The funnel is broken structurally; resume tweaks don't fix it.

Across the candidates we work with, the ones who switched from fifty applications a week to five carefully chosen ones plus a Standout profile saw response rates jump from 2-3% to 25-40%. The math actually moves once you switch lanes.

Verdict

If you're a senior software engineer in or moving to San Francisco, the answer is matching, not applying. The compensation ceiling is the highest in tech, the lab tier is hiring at unprecedented rates, and the application channel is the worst it's been in years. Build a Standout profile. Pick three companies and write to specific engineers there. Spend the rest of your time shipping public artifacts that prove how you think. That's the playbook that converts in May 2026. Period.

If you're entry-level (0-2 years), apply directly to big tech new-grad programs and the structured pipelines at the scale-up tier — those are still optimized for application volume. For everything else, the matching model wins.

FAQ

How many software engineer jobs are there in San Francisco?

Around 14,000 active postings across LinkedIn, Indeed, and direct company sites as of May 2026. The real hiring volume — including sourced and warm-intro hires that never appear as postings — is meaningfully higher. The published number undersells the actual hiring market.

What's the average software engineer salary in San Francisco?

Levels.fyi reports a median total comp of $272,550 across SF software engineers, with the 75th percentile at $375K. Indeed shows base salary averaging $176,198, which is base only and undersells total comp at the AI-native scale-ups and big tech. AI engineering and lab roles run another 50-100% above the general SWE median.

Which companies are hiring the most software engineers in San Francisco?

Anthropic, OpenAI, Google DeepMind, Meta, and xAI lead the AI-native and frontier-lab tier. Stripe, Databricks, Anduril, Ramp, Notion, Figma, and Vercel lead the high-velocity scale-up tier. Big tech (Google, Meta, Apple, Microsoft, Amazon, Salesforce) is steady but slower. The published postings undercount actual hiring across all of these.

Should I move to San Francisco for a software engineering job in 2026?

If you're going to take a senior role at the AI-native scale-up tier or the frontier labs, yes. The comp premium and the equity upside are concentrated in SF and not replicated anywhere else. For mid-tier or generalist enterprise roles, the math is closer — remote-US options now compete on base but not on equity. Run the comp comparison carefully before you sign.

How long does the SF software engineer job search take?

Cold-application path: 3-6 months for senior roles, often longer in current conditions. Sourced or matched path: typically 4-12 weeks to first offer. The compression comes from skipping the application funnel and getting in front of hiring managers directly. The matching model is structurally faster because the company has already self-qualified by the time you hear from them.