Data Scientist Salary Remote (US) — 2026 BLS Data
Salary distribution
Percentile breakdown of Data Scientist base salaries in Remote (US).
A remote data scientist role in 2026 is one of the few seats in the US labor market where the gap between a “fine” offer and a great one can easily exceed $100K — even when both jobs have the same title, the same JD, and the same Zoom interview loop. The reason is geo-banding. The same SQL skills, the same causal-inference toolkit, the same dashboards — priced four different ways depending on which tier a company puts you in.
This guide pulls together what the numbers actually look like in May 2026, where the bands come from, and how to negotiate inside a system that was designed to keep you quiet about it.
The number most recruiters quote you
The BLS May 2024 OEWS release (the most current dataset still in play for 2026 budget cycles) puts the national median wage for data scientists at $112,590, with the 90th percentile at $194,410 and the bottom 10% under $63,650. That is the number that ends up in HR comp surveys, Glassdoor “averages,” and most enterprise job postings.
It is also, for a senior remote data scientist with five-plus years of experience, almost completely useless.
BLS captures every data scientist title in the country — including analytics roles at regional banks, insurance carriers, and state agencies. It does not separate IC3 from IC6, base from total comp, or onsite from remote. Layer in the fact that ~40% of US data science roles are now remote-first or remote-eligible, and the BLS median undershoots what a remote IC at a venture-backed software company actually clears.
What remote DS actually earns in 2026
Pulling Levels.fyi self-reported comp from the last 12 months for US-based remote data scientist roles, the picture looks more like this:
- p25 — $125K total comp: junior IC at a remote-friendly mid-market SaaS, or a senior IC at a non-tech company (retail, logistics, healthcare).
- p50 — $170K total comp: senior IC at a Series C–D software company, fully remote, base around $170K with a small bonus and modest equity.
- p75 — $225K total comp: staff-level IC at a public software company that bands remote at “Tier 2” (more on that below) or a senior IC at an AI lab.
- p90 — $320K+ total comp: staff or principal IC at Meta, Google, or Stripe on remote — possible but increasingly rare in 2026 as Big Tech has rolled back universal remote bands.
The Levels.fyi aggregate p50 for the data scientist title across all companies sits at $177K, which is consistent with the remote-only p50 above once you strip out the non-tech long tail.
How geo-banding actually works
Most US tech employers in 2026 run one of three pay structures:
1. Flat national bands (the GitLab model, the dbt Labs model, the early-Vercel model). One band per level, applied to every US employee regardless of location. GitLab publishes its formula openly in its public handbook — base is calculated from a SF benchmark multiplied by a location factor, but for fully remote US employees, that factor is increasingly flat. Glassdoor and Comparably both peg GitLab’s data scientist base around $138K with $28K bonus and meaningful equity, for ~$167K total at the median IC level.
2. Tiered geo bands (the Stripe model, the Atlassian model, the Snowflake model). Companies bucket US zip codes into 2–4 tiers. Tier 1 is SF, NYC, Seattle, LA. Tier 2 is Boston, DC, Austin, San Diego, Boulder. Tier 3 is most secondary metros. Tier 4 is everything else. Tier 4 base is typically 80–90% of Tier 1. Same role, same level, same JD.
3. Cost-of-labor adjusted on hire (the Meta post-2023 model). Comp is set when you sign based on your home zip code. If you move, your comp can be adjusted downward at your next refresh. This is the model that produced the famous 2022–2023 wave of “if you move out of SF, we cut your pay 15%” headlines — and it is still alive in 2026 at most FAANG-equivalent companies.
The practical takeaway: if you are negotiating remote, the band the recruiter quotes you reflects the cheapest tier they think they can get away with for your zip code. That is not the same as “the band for this role.”
Companies that pay remote DS well in 2026
A few remote-first reference points worth knowing:
- GitLab — fully distributed across 65+ countries, public compensation calculator, data team historically heavy on dbt, BigQuery, and Sisense. Senior DS base lands $140–170K, with bonus and equity bringing total to $165–200K for most ICs.
- dbt Labs — analytics-engineering–native company, hires DS that can read warehouse models directly. Talent acquisition pages reference national bands rather than tiered geo bands as of 2026.
- Vercel — the data team is small but well-paid; the company is one of the few that has held flat national bands through the 2024–2025 down market.
- HashiCorp, Elastic, MongoDB — public remote-friendly software companies with senior DS bands in the $175–230K total comp range.
- Stripe, Notion, Anthropic, OpenAI — pay top of market on remote but apply tiered geo bands. Total comp at staff level routinely crosses $300K.
If your priority is predictability of comp regardless of where you live, the flat-band shops (GitLab, dbt Labs, Vercel) are the cleanest play. If your priority is absolute number and you live in a Tier 1 metro, the geo-banded shops will out-pay them by 20–40%.
Base, bonus, equity — what the mix looks like
For a remote senior data scientist in 2026, a typical $170K-total-comp offer breaks down roughly as:
- Base — $170K (sometimes pulled to $150K to fund a larger equity grant at earlier-stage companies)
- Bonus — $20K (10–15% target, paid annually, prorated in year one)
- Equity — $28K/year (4-year vest of a ~$110K grant at a Series C–D company; less at later-stage public companies, dramatically more at pre-IPO AI labs)
The bonus number is the one that most job seekers under-weight in side-by-side offer comparison. A 10% bonus that pays out at 100% achievement is real money — a “discretionary” bonus that has historically paid 40–60% of target is not. Ask for the last three years of payout history on the bonus pool, in writing. If a recruiter dodges, treat the bonus as zero in your calculus.
Cost-of-living, taxes, and the “remote arbitrage” math
The classic remote arbitrage pitch — “make NYC money in Boise” — is more nuanced in 2026 than it was in 2021. State income tax in Idaho, Tennessee, Florida, or Texas can swing your take-home by 5–9% compared to California or New York. But health insurance markets, property tax, and the actual cost of a 2-bedroom apartment in “low-cost” metros have all risen sharply since 2022. Boise, Austin, and Bozeman are no longer the deals they used to be.
For a flat-band company (GitLab, dbt Labs), arbitrage is real and uncomplicated: same comp, lower COL, higher take-home. For a tiered-band company, the math is more delicate — moving from Tier 1 to Tier 3 can cut your base 15% even if your COL only drops 8%.
Negotiation tactics that work in remote DS interviews
- Anchor on total comp, not base. Recruiters love to lead with base because it is the most “moveable” line item for them. Force the conversation to OTE + equity from the first call.
- Ask for the band, not the offer. “What is the published band for this level at my location?” is a fair question at any company with formal leveling. The recruiter does not have to answer in writing, but the verbal answer tells you the ceiling.
- Get the equity strike price and recent 409A. A “$200K equity grant” at a pre-IPO company is meaningless without the strike price and the most recent 409A valuation. Ask for both.
- Negotiate the band, not just your spot in it. If your skill set genuinely sits at the staff level and you are being offered senior, push the leveling conversation before the comp conversation. The delta between senior and staff at most remote DS shops is $40–80K.
- Use competing offers as range anchors, not threats. Recruiters know within $5K what their competitors are paying for the same role. A competing offer screenshot from Levels.fyi, with the company name redacted, is more persuasive than “I have another offer.”
Tracking offers across remote-first companies
A remote DS job search in 2026 routinely runs 15–30 active applications across a 6-month window — flat-band, tiered-band, public, private, AI lab, fintech, healthtech. Keeping the band data, the recruiter notes, the equity strike prices, and the bonus structures straight across that many companies is its own job.
OfferFlow’s kanban board and salary tracker is built for this exact problem: every job card holds the band, the offer breakdown, the recruiter’s comp range, and the equity terms in one place, so when you have to compare two offers at the close, the math is already done.
Bottom line
The 2026 remote data scientist market is paying — but the band you end up in is downstream of the leveling and geo-band conversations you have in the first two recruiter calls, not the salary number you negotiate at the end. Walk into those conversations knowing the BLS median ($112,590) is the floor, the Levels.fyi p50 ($170K total) is the realistic middle, and the flat-band employers (GitLab, dbt Labs, Vercel) are where remote pay does not get clawed back the day you move.