What you actually want to know is more specific: How much will I make? In which role? At what stage of my career? In which city? And what’s the realistic ceiling not the LinkedIn brag post ceiling, but the real one?
Those are fair questions. And the honest answer is that “computer science salary” is one of the most misleading phrases in career research, because it bundles together roles that pay $58,000 and roles that pay $340,000 and presents them as if they exist on the same spectrum. They don’t. They’re practically different professions.
So here’s what I’m going to do. I’m going to break down actual compensation by role, level, and geography with real numbers, real caveats, and real talk about what the averages are hiding. By the end of this, you’ll have a clear map of what a computer science career actually pays across the major paths, and more importantly, you’ll understand why the numbers land where they do.
First, Why Most CS Salary Data Is Misleading
Before we get into numbers, we need to talk about methodology for about 60 seconds — because if you don’t understand how salary data gets collected, you’ll misread every number that follows.
Most salary surveys (Glassdoor, LinkedIn Salary, PayScale) are self-reported. Someone fills out a form, chooses their job title from a dropdown, and enters their comp. The problems with this are obvious: titles aren’t standardized across companies, people round up, and the sample skews toward workers who are actively job hunting or recently hired — meaning they’re not a representative cross-section of the entire profession.
The BLS collects employer-reported data, which is more reliable but lags reality by 12–18 months and doesn’t capture equity or bonuses — which at senior levels can double or triple base salary.
Levels.fyi collects verified compensation data specifically from software engineers at tech companies, breaking out base, bonus, and equity. It’s the gold standard for the tech industry specifically, though it skews toward large tech companies and underrepresents smaller firms and non-tech industries.
Here’s what that means practically: when you see “average software engineer salary: $120,000,” that number is simultaneously too low for a senior engineer at Google and too high for a junior developer at a 30-person startup in Omaha. Both workers have “software engineer” on their LinkedIn. Neither is lying.
With that caveat firmly established, let’s get into the actual numbers.
Software Engineer / Software Developer
This is the role most people picture when they hear “computer science career,” and it’s also the role with the widest compensation range of any CS path. Which makes it both the best and worst benchmark.
By experience level (U.S. national median, base salary):
Entry-level (0–2 years): $85,000–$105,000
Mid-level (3–5 years): $115,000–$145,000
Senior (6–10 years): $155,000–$195,000
Staff / Principal: $200,000–$270,000
Distinguished / Fellow: $280,000–$400,000+
Those numbers are base salary only. At large tech companies — your FAANGs, your Microsofts, your Salesforces — total compensation at the senior and above levels often looks radically different once you add in annual bonuses (typically 10–20% of base) and RSU (restricted stock unit) vesting. A senior engineer at Meta with a $195,000 base might have a total compensation package worth $320,000–$380,000 when equity is included.
At a mid-size startup, that same senior engineer might earn $160,000 base with meaningful but unproven equity. At a mid-size company outside the tech sector — a bank, a hospital system, a manufacturing firm with an IT department — the same title might pay $130,000–$145,000 with a traditional bonus structure and no equity.
Same title. Three wildly different comp structures. This is why “what does a software engineer make?” is almost an unanswerable question without context.
What actually drives the spread: The single biggest factor in software engineering compensation isn’t years of experience. It’s the type of company and the industry. A software engineer at a hedge fund or high-frequency trading firm will consistently out-earn their counterpart at a traditional enterprise company by 40–70%, even with identical experience and skills, because financial services companies monetize engineering directly.
Data Scientist
Data science sits in an interesting spot in the CS salary landscape it pays well, but the ceiling is lower than software engineering at top companies, and the floor is surprisingly lower than many candidates expect.
By experience level (U.S. national median, base salary):
Entry-level (0–2 years): $90,000–$110,000
Mid-level (3–5 years): $120,000–$150,000
Senior (6–10 years): $155,000–$185,000
Principal / Lead: $190,000–$240,000
Here’s something the “data science is the sexiest job of the 21st century” narrative doesn’t tell you: the market for data scientists bifurcated hard between 2020 and 2024. On one side, there’s strong demand for data scientists who can build production ML systems, work with large-scale data infrastructure, and communicate quantitative findings to business stakeholders. Those people are well-compensated and employable.
On the other side, there’s a glut of candidates who completed data science bootcamps or online courses, can run a Jupyter notebook, and describe themselves as data scientists. The entry-level market for this second group is considerably more competitive than it was in 2019–2021, and starting salaries have compressed in many markets.
The honest breakdown of data science comp comes down to one question: are you doing analytical data science (dashboards, business intelligence, reporting, exploratory analysis) or machine learning engineering (building and deploying models in production systems)? The former typically pays 15–30% less than the latter. If you’re doing ML engineering work, you’re increasingly being hired under ML Engineer titles anyway.
What accelerates earnings in data science: Domain expertise. A data scientist who deeply understands healthcare systems, financial markets, or supply chain logistics not just the technical methods commands significant premiums. Generalist data scientists are a commodity. Domain-specialist data scientists are not.
Machine Learning Engineer / AI Engineer
This is the highest-paying role in computer science right now. Not because ML engineers are smarter than other engineers — they’re not, or at least that’s not the variable — but because supply is dramatically lower than demand, and the economic value companies extract from working ML systems is enormous.
By experience level (U.S. national median, base salary):
Entry-level (0–2 years): $120,000–$145,000
Mid-level (3–5 years): $160,000–$200,000
Senior (6–10 years): $210,000–$260,000
Staff / Principal: $270,000–$350,000+
At AI-focused companies — OpenAI, Anthropic, DeepMind, Google DeepMind, Meta AI — total compensation packages for senior ML researchers and engineers have been reported well above $400,000, with some principal-level roles exceeding $600,000–$800,000 when significant equity grants are included. These are outliers, but they’re real outliers.
The reason for this premium comes down to supply constraints. ML engineering requires a combination of strong software engineering fundamentals, linear algebra and calculus-level mathematics, deep familiarity with ML frameworks (PyTorch, JAX, TensorFlow), experience with distributed computing and large-scale data pipelines, and increasingly, hands-on experience with transformer architectures and large language model fine-tuning. The overlap of all those skills in a single person is genuinely rare.
What’s happening in 2025 is that the AI investment supercycle is generating more ML engineering job postings than qualified candidates exist to fill them. According to Indeed data, ML engineer postings increased 74% year-over-year between 2023 and 2024, while the qualified candidate pool grew at a fraction of that rate.
If you’re in a position to develop these skills and you have a solid CS or mathematics foundation the compensation premium is as real as any in the profession right now.
Cybersecurity Engineer / Information Security Analyst
Cybersecurity is the most reliably employable path in computer science, full stop. It’s not always the highest-paying path at the entry level — but it has the lowest unemployment rate, the most consistent demand across industries, and arguably the clearest skills-to-certification-to-employment pipeline of any CS specialty.
By experience level and role (U.S. national median, base salary):
Security Analyst (entry, 0–2 years): $70,000–$90,000
Security Analyst (mid, 3–5 years): $100,000–$130,000
Security Engineer (mid-senior): $130,000–$165,000
Penetration Tester / Red Team: $120,000–$170,000
Cloud Security Architect: $160,000–$210,000
CISO (Chief Information Security Officer): $195,000–$350,000+
The BLS projects 32% job growth for information security analysts through 2033 — the highest projected growth rate of any CS occupation. CyberSeek’s 2024 workforce analysis found over 460,000 unfilled cybersecurity positions in the U.S., with the average position remaining open for 21 weeks.
That gap between open positions and available talent is the structural reason cybersecurity compensation has risen consistently even during the 2022–2023 tech layoff cycle that hit software engineering hard. Banks, hospitals, defense contractors, government agencies, and virtually every other large organization need security talent — and it’s not concentrated in the Bay Area the way software engineering compensation is.
The caveats: entry-level cybersecurity is harder to break into without certifications or a CS degree than many people expect. CompTIA Security+ is the baseline credential most employers recognize; CISSP is the gold standard for senior roles. The “I watched YouTube tutorials on ethical hacking” pipeline is real but slow. Formal credentials accelerate it significantly.
DevOps / Cloud Engineer / Site Reliability Engineer
DevOps and its related disciplines SRE, platform engineering, cloud architecture — occupy a fascinating salary position: consistently paid at or above software engineering levels, but significantly less discussed in the popular narrative about CS careers.
By experience level (U.S. national median, base salary):
Entry-level DevOps / Cloud Engineer: $90,000–$115,000
Mid-level: $130,000–$160,000
Senior DevOps / SRE: $165,000–$200,000
Principal SRE / Platform Architect: $210,000–$270,000
Cloud Architect (enterprise): $175,000–$240,000
The cloud certification market AWS, Google Cloud, Azure — has produced a large pool of credential-holders. But there’s a meaningful gap between someone who passed the AWS Solutions Architect Associate exam and someone who can design a multi-region, fault-tolerant infrastructure for a system handling millions of requests per hour. The latter earns considerably more.
SRE (Site Reliability Engineering) is worth specific mention because it’s the highest-compensation track in this category and the least understood. Google invented the SRE model it’s essentially software engineering applied to operations, with a heavy emphasis on reliability, scalability, and incident management. Google, Netflix, Stripe, and similar companies pay SREs at or above senior software engineer rates, because the cost of downtime at scale is enormous and the people preventing it are genuinely hard to find.
Data Engineer
Data engineering doesn’t get the same cultural cachet as data science or ML engineering, but in terms of employment stability and compensation trajectory, it’s one of the strongest bets in CS right now.
By experience level (U.S. national median, base salary):
Entry-level (0–2 years): $90,000–$115,000
Mid-level (3–5 years): $125,000–$155,000
Senior (6–10 years): $160,000–$195,000
Principal / Staff Data Engineer: $200,000–$250,000
Data engineering — building the pipelines, infrastructure, and systems that make data science and business analytics possible — has become a critical function at every data-driven organization. Every company that wants to “do AI” first needs reliable, clean, accessible data. The people who build those systems are data engineers.
What’s driven compensation growth here specifically: the transition from traditional ETL pipelines to modern data stack tools (dbt, Airflow, Spark, Snowflake, Databricks) created a skills gap between experienced engineers who know legacy systems and engineers who can build modern data infrastructure. That gap is narrowing, but it’s still real enough to create premiums for people who can work fluently across both worlds.
Product Manager (Technical)
Wait is product management a computer science job? Formally, no. Compensation-wise, it absolutely competes.
By experience level (U.S. national median, base salary):
Associate PM (0–2 years): $100,000–$125,000
PM (3–5 years): $140,000–$175,000
Senior PM (6–10 years): $180,000–$220,000
Director of Product: $220,000–$290,000
VP of Product: $280,000–$400,000+
Technical product managers people who manage software products, work directly with engineering teams, and can credibly engage with technical architecture decisions — consistently command higher compensation than non-technical PMs. A CS degree isn’t required for the role, but it’s an enormous advantage, and many tech companies explicitly prefer or require it.
The equity component in PM compensation at growth-stage startups is where the truly large outcomes come from. A senior PM at a Series B startup with meaningful equity who hits a good liquidity event can see total realized compensation that dwarfs almost any other CS-adjacent path. It’s also speculative in a way that a salary is not — so factor that into your risk profile accordingly.
IT Support / Help Desk (The Real Entry Floor)
Let’s be honest about where the floor is, because career content has a tendency to skip it.
By experience level:
Tier-1 Help Desk / Support Specialist: $42,000–$62,000
Tier-2 Technical Support: $58,000–$78,000
Systems Administrator (entry): $65,000–$85,000
Senior Systems Administrator: $90,000–$115,000
These roles are classified as IT. They are not, typically, computer science careers in the sense of applying CS principles and algorithms. They’re operational technology roles keeping systems running, troubleshooting end-user issues, managing software deployments.
The reason I’m including them: a meaningful number of people with CS degrees end up here initially, especially in smaller markets or without internship experience. There’s nothing wrong with starting here. The question is whether you’re using the role as a launchpad — studying for certifications, building projects, developing skills that move you toward the higher-compensation tracks — or settling.
The people who spend their entire career in help desk and systems administration and never develop beyond it are systematically underpaid relative to the broader CS market. The people who use operational IT as a foundation while building toward cloud, security, or engineering roles often find it was a useful first chapter.
The Geography Factor (This Is Bigger Than Most People Admit)
National medians are real, but they flatten a variance that genuinely matters to your life.
Software Engineer, Senior — Same title, different markets (total comp, base + typical bonus):
San Francisco Bay Area: $220,000–$320,000
New York City: $195,000–$280,000
Seattle: $190,000–$270,000
Austin, TX: $155,000–$210,000
Chicago: $145,000–$195,000
Atlanta: $140,000–$185,000
Denver: $145,000–$190,000
Columbus, OH: $120,000–$155,000
Remote (company-dependent): $140,000–$230,000 (varies enormously by company policy)
The Bay Area premium is real but has compressed since 2020. Remote work expanded the viable talent pool for tech companies, which increased competition for remote talent and pushed salaries in secondary markets up — while simultaneously giving tech workers the option to take Bay Area salaries without Bay Area cost of living.
The catch: many large tech companies have implemented location-adjusted remote pay. At Google, Meta, and several other majors, accepting a full-remote position in Columbus, Ohio pays 15–25% less than the same role based in San Francisco, even if you’re doing identical work. Some companies have no location adjustment. It varies widely, and it’s worth explicitly understanding before you accept an offer.
The places where geography creates the biggest career leverage: cybersecurity in the Washington D.C. metro (defense and government contractors drive extraordinary demand), financial technology in New York City, aerospace and defense tech in Seattle and Southern California, and healthcare IT in markets like Nashville, Cleveland, and Minneapolis.
What the Comp Charts Don’t Show: Equity, Benefits, and Real Total Compensation
We’ve been talking about base salary and noted bonus and equity. But let me be specific about why this matters, because the gap between base salary and total compensation is genuinely enormous at certain companies and essentially zero at others.
At large public tech companies (Google, Meta, Apple, Microsoft, Amazon): Expect RSU grants that vest over 4 years. A senior engineer offer might be $185,000 base + $30,000 annual bonus + $800,000 RSU grant vesting over 4 years = $200,000/year from equity alone at grant price. Total annual comp: $415,000. The base salary on that offer? $185,000. If you benchmark against base, you’re comparing apples to tractors.
At growth-stage startups (Series B–D): Equity comes as options, typically at a strike price set during a 409A valuation. The value is entirely speculative until a liquidity event. Some of those options become life-changing. A meaningful percentage become worthless. That’s the risk you’re taking in exchange for potentially a below-market base salary and a higher upside.
At non-tech companies (banks, hospitals, manufacturers, retailers): Equity is often nonexistent, replaced by profit-sharing or traditional pension programs. The total compensation picture may be more stable but the ceiling is typically lower. The trade-off is job security, work-life balance, and predictability — which genuinely matter depending on where you are in your life.
Benefits worth money: Health insurance premiums (Google and Meta fully cover premiums, saving $10,000–$20,000/year vs. paying market rates), 401(k) matching (typically 3–6% of salary), education reimbursement ($5,000–$25,000/year at major companies), and in some cases housing stipends or relocation packages. These aren’t trivial they’re real compensation.
The Honest Career Advice Hidden in These Numbers
Here’s what the salary data is actually telling you, if you read it right.
Specialization pays more than generalization, with one exception. Specialized CS roles — ML engineer, cloud security architect, SRE — consistently out-earn generalist roles. The exception is a truly exceptional generalist (usually a small startup CTO or principal engineer), who commands a premium for breadth. Early in your career, developing depth in one area matters more than trying to know everything.
The company type matters as much as the role. A mid-level data scientist at a hedge fund or quantitative finance firm often out-earns a senior data scientist at a traditional enterprise company. Industry matters. Finance, tech, and defense pay more. Healthcare IT, education tech, and nonprofit tech pay less. That’s structural, not personal.
The 10-year trajectory matters more than year-one salary. An entry-level ML engineer at $130,000 in a role with real growth opportunity will almost certainly earn more over a decade than a starting salary of $105,000 that plateaus at $140,000 because there’s no pathway to principal-level work. Think in trajectories, not snapshots.
Remote work changed the map but didn’t flatten it. You can now access Bay Area-tier compensation without living in the Bay Area — but only if you’re competitive enough to get those jobs. The companies offering premium remote compensation are also typically the most competitive to get into. Location flexibility is real. It just doesn’t come free.
Quick Reference: CS Salaries by Role (2025)
| Role | Entry Level | Mid Level | Senior | Principal/Staff |
|---|---|---|---|---|
| Software Engineer | $85K–$105K | $115K–$145K | $155K–$195K | $200K–$270K |
| ML / AI Engineer | $120K–$145K | $160K–$200K | $210K–$260K | $270K–$350K+ |
| Data Scientist | $90K–$110K | $120K–$150K | $155K–$185K | $190K–$240K |
| Data Engineer | $90K–$115K | $125K–$155K | $160K–$195K | $200K–$250K |
| Cybersecurity Engineer | $70K–$90K | $100K–$130K | $130K–$165K | $175K–$240K |
| DevOps / SRE | $90K–$115K | $130K–$160K | $165K–$200K | $210K–$270K |
| Technical PM | $100K–$125K | $140K–$175K | $180K–$220K | $220K–$290K |
| IT Support / SysAdmin | $42K–$62K | $65K–$85K | $90K–$115K | $115K–$140K |
Base salary, U.S. national, 2025. Total compensation at large tech companies is typically 40–100% higher when equity and bonuses are included.
The Bottom Line
Computer science is a field where your salary is genuinely determined by the intersection of role, company type, geography, and the specific skills you’ve developed — not by the degree itself. The CS degree is the entry ticket. What you do with it determines where you land on a spectrum that runs from $58,000 to well over $400,000.
The roles paying the most right now — ML engineering, AI infrastructure, cloud security architecture — all reward a specific combination of deep technical fundamentals and domain-specific expertise that takes real time and focus to develop. The salary premium exists because the supply of people who’ve done that work is genuinely limited.
The most important number in any of these ranges isn’t the median. It’s the ceiling and understanding what gets you there.
