[Salary Data Updated in October 2026]
A data scientist in India earns ₹14,00,000 per year on average, according to Glassdoor India's total pay figure as of 16 October 2026. Total pay here means fixed salary plus bonus and any other cash component, so it sits above the base-only numbers you will find quoted elsewhere.
What makes that number interesting in 2026 is not how fast it has moved, but how little the AI hiring boom has moved it. An AI engineer in India averages ₹13,50,000, and a machine learning engineer averages ₹14,00,000, putting all three titles within ₹50,000 of each other. The premium you keep reading about, usually quoted somewhere between 25 and 40 percent, doesn't appear in the middle of the distribution.
It appears at the edges instead. The highest package reported for a data scientist is ₹21,05,000, while AI engineers report up to ₹24,12,500. The bonus component tells the same story: AI engineers see additional pay stretch to ₹5,00,000, compared with ₹3,00,000 for data scientists. In other words, AI is widening the top of the range and loading more of the package into variable pay, rather than lifting the floor for everyone who adds an AI skill to their resume.
Before the details, a quick definition, because the breadth of the role is exactly why the salary band is so wide.
Data science turns raw data into decisions a business can act on, drawing on statistics, programming, and domain knowledge at the same time. Most projects move through four stages: you ingest data from databases, APIs, or sensors; clean and store it; analyze and model it; then present the results so someone can decide something.
Your pay tracks how much of that chain you actually own. Someone who only runs the analysis stage and hands a chart to a manager earns very differently from someone who frames the question, builds the model, ships it, and then defends the trade-offs to the business. This article explores the four variables that determine where you land: job title, city, employer, and the depth of your technical skills.
What Is the Average Data Scientist Salary in India?
Glassdoor splits every package into two parts, and the split matters more than most salary articles admit. Base pay is your fixed monthly salary, the amount you can plan your rent around. Additional pay covers performance bonuses, joining bonuses, and other cash incentives, meaning it can disappear in a bad year.
Source: Glassdoor India, Data Scientist salaries, accessed 16 October 2026. Figures are self-reported by employees and are recalculated by Glassdoor over time.
Look at how wide the base pay band is before anything else: for people with the same job title in the same country, the top pays more than twice the bottom. That spread is the single most useful fact on this page, because it tells you that the title on your offer letter explains far less about your salary than most candidates assume.
Why Salary Figures for the Same Role Disagree So Widely?
If you search this question across five websites, you will get five different answers, often 20 or 30 percent apart. It is worth understanding why rather than assuming one of them is lying. There are four structural reasons for this. Once you know them, you can read any salary figure properly.
Aggregators report self-submitted data: Glassdoor's India figure for data scientist rests on roughly 19,500 anonymous submissions collected over several years, then weighted by a prediction model that recalculates continuously. Two loads of the same page on the same day can return different numbers, which is why every figure on this page carries a date.
The base-versus-total confusion described above: A site quoting ₹12,00,000 and a site quoting ₹14,00,000 may be looking at exactly the same Indian data scientist, with one reporting base pay and the other reporting base plus bonus. Neither is wrong, but comparing them directly is.
Title drift is the big one in India, specifically. "Data scientist" is stamped on a reporting analyst building dashboards at an IT services firm and on a modeling specialist shipping recommendation engines at a product company. As a result, the title became a hiring magnet, and employers applied it loosely. When one job title covers two genuinely different jobs, the reported band has to be wide enough to hold both.
Sample composition: If more employees from volume-hiring services firms submit salaries in a given quarter, the national average drops, even though nobody took a pay cut. The reverse happens when product-company employees submit in bulk.
So when you find a figure, ask three questions before you use it in a negotiation. Is this base or total pay, when was it measured, and does it cover India or a single city? A number without those three facts attached cannot tell you anything useful about your own offer.
Data Scientist Salary Per Month in India
Annual figures are the industry convention, but they hide what actually arrives in your account each month, so here is the same data divided by twelve.
Monthly figures are calculated from the Glassdoor India annual averages above, divided by twelve.
Treat these as gross figures, because income tax, provident fund, and professional tax are deducted before the money reaches you, so your in-hand amount will be meaningfully lower. The bonus component also does not spread evenly across twelve months: it usually lands once a year or once a quarter, which means your actual monthly inflow is closer to base pay divided by twelve than to the numbers above.
Data Scientist Salary by Job Title in India
Because of the title drift problem, job title predicts your pay better than years of experience does, and the ladder below is the clearest view of the market you can get from a single source. Two people with six years each can sit two levels apart and earn very different amounts, purely because one got promoted into a senior title and the other did not.
Source: Glassdoor India, Data Scientist salaries, accessed 16 October 2026. Figures are self-reported by employees and are recalculated by Glassdoor over time. Glassdoor reports by job title, not by years of experience. The trajectory band is a separately modeled estimate and does not reconcile exactly with the average total pay column, so read the two as different measures rather than as a range around a midpoint.
The promotion from data scientist to senior data scientist is the steepest financial event in this entire career, because average total pay doubles across that single step. Nothing else on the ladder comes close, which is why the middle section of this article spends most of its time on what that promotion actually requires.
After that point, the curve flattens noticeably. Lead and senior share the same ₹28,00,000 average, but lead stretches the top of the band to ₹35,00,000 rather than raising the middle. Moving from senior to lead therefore improves your best case more than your expected case, which is a useful thing to know before you chase the title.
The director band looks almost broken at ₹14,00,000 to ₹68,00,000, and that width is itself informative. It mixes someone running a three-person analytics team at a mid-sized company with someone owning an enterprise data function, and both legitimately carry the title of director.
If you are working toward that first big promotion, the Data Specialization Course is built around the end-to-end project work that senior interviews actually probe.
Data Scientist Salary by City in India
City moves your package by roughly ₹4,00,000 at the same title, and the reason is less about cost of living and more about which employers happen to cluster where.
Source: Glassdoor India, Data Scientist salaries, accessed 16 October 2026. Figures are self-reported by employees and are recalculated by Glassdoor over time.
Gurgaon tops the list, not Bengaluru, which surprises most people. This is because of what we call the employer mix. Consulting firms, global capability centers of foreign banks, and fintech employers cluster in that corridor, and they compete for the same relatively small pool of candidates who can model risk and explain it to a client.
Hyderabad sits lowest of the six for the mirror-image reason. Its data science hiring leans toward captive delivery centers of large IT services firms, which hire in volume against fixed internal pay grades. Volume hiring anchors an average downward no matter how much the city grows.
Glassdoor models its national average and its city averages separately, so the ₹14,00,000 national figure sits above five of the six cities listed here, which cannot be literally true. Treat the city column as a reliable guide to how these markets rank against each other, and not as numbers that average out to the national figure.
Read all of these as gross figures rather than spending power, though. Rent and commuting in Gurgaon and Mumbai absorb much of the premium over Pune or Hyderabad, so a ₹12,05,000 package in Pune can leave you better off month to month than a ₹14,95,000 package in Gurgaon.
Data Scientist Salary by Company in India
If city is really a proxy for employer mix, then employer type is the variable doing the actual work, and the table below shows it directly.
Source: Glassdoor India, Data Scientist salaries, accessed 16 October 2026. Figures are self-reported by employees and are recalculated by Glassdoor over time.
The gap between the top and bottom of this table is ₹10,00,000 for the same job title, and it maps almost perfectly onto one thing: whether data science is the product or a service line.
At Fractal and Tiger Analytics, analytics is what the company sells, so a modeling hire connects directly to billable value. So, the firm competes hard for strong candidates. At a broad IT services firm, companies hire data scientists at scale across many client accounts and slot them into existing engineering pay grades, which caps what any individual can command.
This has a practical consequence that matters more than any city or title decision you will make. The median data scientist at Fractal earns ₹10,00,000 more than the median data scientist at Tata Consultancy Services for the same title and years of experience. This gap is larger than any single promotion on the ladder, except the jump to senior. Switching employer type is therefore often the faster route to a higher number than waiting for a promotion where you are.
Some teams can also sit far above their company's national median, so check that before you rule an employer out. Mastercard in Gurgaon reports a ₹24,00,000 median for the data scientist title and Amazon in Delhi reports ₹20,00,000, both well above the national average.
Note that employer-level figures are measured per company, so they can and do exceed the highest package reported in the national dataset.
The Data Scientist Career Path in India
Everything above describes where the destinations sit. This section is about the route between them, what changes in the work at each stage, and what you can do now to move faster.
Glassdoor reports pay by job title rather than by years of experience, so the year ranges below reflect common Indian hiring practice rather than reported data, and your own timeline may run faster or slower.
Years 0 to 3: Learning to Finish Things
In your first few years, you work inside problems that other people have already framed, which means exploratory analysis, feature engineering, and model validation under somebody else's review. Pay sits in the ₹7,00,000 to ₹18,00,000 band, and where you land inside it depends far more on employer type and portfolio strength than on your degree.
The single behavior that separates people who move up quickly from people who stall is finishing. A model that reached production and is still running tells an interviewer something that five polished notebooks cannot, because it proves you handled the unglamorous parts: the missing data, the retraining, the monitoring, the thing that broke at 2am. Owning the SQL and Python work end to end, rather than inheriting clean datasets, does the same job.
If you are building that foundation now, our Data Science using Python covers the Python and SQL layer these roles screen for first.
Years 4 to 7: Where the Pay Doubles
This is the promotion that matters, the one where average total pay moves from ₹14,00,000 to ₹28,00,000. It is worth knowing why employers will pay twice as much. The work changes in kind, not just in scale: you now frame the problem yourself, choose the approach, and defend the trade-offs to people who will never read your code.
PS: Employers screen hardest for the second row in this table.
What changes | What it looks like in practice |
|---|---|
Problem framing | You translate a vague business ask into a measurable modeling problem |
Production ownership | You handle deployment, monitoring, and retraining, not just training |
Stakeholder work | You defend model choices to people who do not read code |
Review duty | You review the work of junior data scientists |
A model sitting in a notebook doesn't count at this level because the hard engineering and most of the business risk live in everything that happens after training finishes. Candidates who can talk credibly about a model degrading in production, and what they did about it, clear this bar; candidates who can only talk about accuracy scores usually do not.
Years 8 and Beyond: Three Paths to Choose From
The ladder stops being a ladder here, because you choose a direction rather than climbing a step, and the three options pay and behave quite differently.
Lead Data Scientist (₹18,00,000 to ₹35,00,000): you stay hands-on technical and own a workstream plus the handful of people delivering it.
Data Scientist Manager (₹20,00,000 to ₹30,00,000): you shift to headcount, hiring, and delivery accountability, and your own modeling time largely disappears.
Principal Data Scientist (₹30,00,000 to ₹55,00,000): you remain an individual contributor but set technical direction across several teams.
Principal pays the most of the three by a clear margin. It also has the fewest openings, because most organizations run one principal across many teams while every growing data team needs managers. This trade-off is the real decision: the manager route gives up top-end pay in exchange for a much wider job market and far more portable experience.
The Skills Pushing the Top of the Band Upward
Earlier in this article, the AI titles showed a lower average but a higher ceiling than data scientist. That pattern is the most useful signal in the whole dataset for anyone deciding what to learn next. It says that the skills raising pay in India right now are the ones that get models in front of users, not the ones that build better models in isolation.
Title | Average total pay | Highest reported package |
|---|---|---|
Data Scientist | ₹14,00,000 | ₹21,05,000 |
Machine Learning Engineer | ₹14,00,000 | ₹21,00,000 |
AI Engineer | ₹13,50,000 | ₹24,12,500 |
Source: Glassdoor India, Data Scientist salaries, accessed 16 October 2026. Figures are self-reported by employees and are recalculated by Glassdoor over time.
Two things follow from that table.
Application-layer AI work, meaning building products on top of existing large language models rather than training your own, sits behind the higher AI engineer ceiling and the larger bonus component, because that work ties directly to a shipped feature somebody can measure. Deployment and operations skills appear to carry a similar senior premium, though Glassdoor's own pages disagree on the size of it: one puts the senior machine learning engineer band at ₹15,00,000 to ₹34,00,000 and another at ₹14,00,000 to ₹24,00,000 on the same day, so treat the direction as sound and the magnitude as unsettled.
The other skill is the least technical and the most reliably rewarded: framing a business problem and defending a modeling decision to a non-technical audience. That is precisely what analytics-first firms sell, and it is why their medians sit ₹10,00,000 above volume-hiring services firms for the same title. Domain depth in one sector, whether that is credit risk, supply chain, or healthcare, compounds that advantage because the framing gets sharper the better you understand the business.
A fair caution here. Glassdoor does not publish salary by individual skill for India, so nobody can honestly attach a rupee figure to "learning MLOps" or "learning LLM application development". The title and band data support the direction of travel, not a price tag, and you should be skeptical of any article that quotes one.
How to Prepare For Your Next Promotion
Pulling it all together, five things move your number, roughly in order of how much they move it.
Ship one model into production and keep it running, then learn to talk about what broke. This single piece of evidence carries more weight in a senior interview than any certification.
Learn the deployment stack properly, meaning cloud, containerization, monitoring, and retraining, because that is where the senior premium demonstrably sits.
Own the data layer end to end. Candidates who can write the SQL, build the pipeline, and explain the data quality trade-offs are rarer than candidates who can tune a model.
Pick one domain and go deep, so your problem framing is sharper than someone who's touched six industries lightly.
Target employer type deliberately when you switch, because moving from a services firm to an analytics-first firm changes your band more quickly than a promotion will.
For the deep learning and neural network layer that the machine learning engineering route assumes, see Deep Learning with Python, and for the modeling foundation beneath it, see the Machine Learning Course.
Reading resources: How to Become a Data Scientist | Data Scientist vs Machine Learning Engineer | Data Scientist Jobs in Hyderabad
Ready to move up a band? Check Your Eligibility for the Data Science Specialization Course.
FAQs
What is the average salary of a data scientist in India?
Glassdoor India reports an average total pay of ₹14,00,000 per year as of 16 October 2026. That splits into average base pay of ₹12,00,000 and average additional pay of ₹2,00,000. Base pay itself ranges from ₹8,00,000 to ₹18,10,000, which is a wide spread for a single job title.
What is the data scientist salary per month in India?
At the national average of ₹14,00,000 per year, gross monthly pay is about ₹1,16,667. A senior data scientist averages close to ₹2,33,333 per month. Your in-hand amount is lower because income tax, provident fund, and professional tax come out first.
Which city pays the highest data scientist salary in India?
Among major hiring markets, Gurgaon pays the most, with an average total pay of ₹ 14,95,000. Bengaluru follows at ₹13,50,000, then Delhi at ₹13,20,000. Hyderabad ranks lowest of the six at ₹11,00,000, largely because captive IT services centers dominate hiring there.
What is the highest salary for a data scientist in India?
The highest package reported for the data scientist title itself is ₹21,05,000, though individual employers exceed it: Mastercard in Gurgaon reports a ₹24,00,000 median. Senior titles go much further. The principal data scientist band reaches ₹55,00,000, and the director of data science band extends to ₹68,00,000.
Which company pays the highest data scientist salary in India?
Among large Indian employers, Fractal reports the highest median at ₹18,00,000. Tiger Analytics follows at ₹16,00,000 and IBM at ₹15,00,000. Specific city teams pay more. Mastercard in Gurgaon reports a median of ₹24,00,000 for the same title.
Why do data scientist salary figures vary so much between websites?
Four reasons. Aggregators continuously reweight self-submitted data, so the same page can return different numbers. Some sites quote base pay while others quote total pay including bonus. In India, the title also covers very different jobs across employer types, and the reported average shifts with the sample mix.
What is the career path for a data scientist in India?
The path runs from data scientist to senior data scientist, then splits into lead, manager, or principal. The steepest jump is from data scientist to senior, where average total pay doubles from ₹14,00,000 to ₹28,00,000, because the role shifts from executing analysis to framing problems and owning models in production. Principal pays the most of the three senior doors but has the fewest openings.