Interview Sarthi

Home › ApplySarthi › Open-jobs dataset

What 60,975 open jobs in India actually contain

Job-market reports usually count postings. This one counts what is written inside them, because that is what decides whether you can search properly.

First-party dataset · Snapshot taken 18 September 2026 · Free to reuse with attribution

Everyone publishes counts of how many jobs are open. Very little is published about what those postings actually say, which is the part that decides whether a filter works, whether a matcher can read the role, and whether you can tell what a job pays before you apply.

This is a snapshot of every posting ApplySarthi held open on 18 September 2026, taken from its own database. It is published because we have not found the equivalent anywhere else.

The headline finding: four postings in five say nothing about pay

Salary appears in 19.9% of open postings. Not a range, not a band — any mention of pay at all. Years of experience fares barely better at 19.3%, and a structured skills list appears on 18.3%.

The practical consequence is that salary and experience are useless as primary search filters in India. Filtering on either silently discards about 80% of what is open, including most of the jobs you would have wanted. Any product that offers them as headline filters is hiding most of the market from you.

Field coverage across 60,975 open postings

Field	Present on	Notes
Company name	99.0%	Almost always present, though ATS-sourced rows carry the hiring entity, not the brand.
Posting date	99.5%	Reliable enough to filter on. Freshness is the one filter worth trusting.
Description text	96.7%	The body a matcher can actually read.
Location string	96.3%	Present, but often "India" or "Remote" rather than a city.
A resolvable city	41.3%	Fewer than half of postings name a city a filter can use.
Company size	20.9%	Mostly from Wellfound, which publishes it.
Salary	19.9%	Four postings in five say nothing about pay.
Years of experience	19.3%	Stated as a number far less often than job seekers expect.
Skills list	18.3%	Structured skills are the exception, not the rule.
Closing date	5.7%	Almost nobody publishes one; a job is open until it disappears.

Where the jobs come from

By source, the same snapshot. Company career boards — Greenhouse, Lever, Ashby, Workday, Oracle HCM and the large in-house boards — together outweigh any single job board.

Source	Open jobs
Indeed	12,708
Greenhouse	9,796
LinkedIn	8,110
Naukri	6,293
Lever	4,837
Shine	3,452
Ashby	3,053
Workday	2,871
Amazon	2,333
Oracle HCM	1,651
IBM	1,113
Internshala	1,093

Where the jobs are

Among the 41.3% of postings where a city could be resolved at all, the concentration is steeper than most people expect. Bengaluru alone holds more open roles than Delhi NCR, Hyderabad and Mumbai combined.

City	Open jobs
Bengaluru	10,632
Delhi NCR	3,474
Hyderabad	3,254
Mumbai	2,163
Pune	2,040
Chennai	1,662
Ahmedabad	398
Kolkata	309
Kochi	198
Chandigarh	186

Read the long tail carefully before drawing conclusions about it. Ahmedabad, Kolkata, Kochi and Chandigarh look small here partly because they are smaller markets and partly because postings outside the big six more often carry a state or a bare “India” instead of a city, and so fall into the 58.7% with no resolvable city at all.

Freshness

26,700 of the 60,975 open postings — about 44% — were posted within the last seven days. Posting date is present on 99.5% of rows, which makes it the one field reliable enough to filter on hard. A closing date, by contrast, appears on 5.7%: in practice an Indian job posting is open until it disappears, and nobody tells you when that will be.

Method, and what this is not

Figures come from the ApplySarthi database on 18 September 2026: every posting it had collected from 21 sources and 754 company career boards and still believed to be open. It is a large sample of the Indian online job market, not a census of it.

Counts are of open postings, not of vacancies or hires; one vacancy advertised on three sites appears once after deduplication but a genuinely duplicated posting may survive. Coverage percentages measure whether a field was present and non-empty, not whether it was accurate. Sources are weighted by what is technically reachable, so this over-represents employers using public ATS boards and under-represents those advertising only on sites that block collection. It is a large sample of the Indian online job market, not a census of it, and it describes one day.

Questions people ask