The Listings Are Not Where Most Engineers Are Looking
Most engineers run their search on LinkedIn, Indeed, or Glassdoor and assume that is where the volume is. That assumption is wrong. A parallel market of 408,448 active listings sits directly on company career pages, collected through the public APIs of eight applicant tracking systems, and most engineers never touch it. The listing count is understated for 16 employers due to a collection ceiling of 2,000 listings per company on one platform; the employer counts are not affected by that ceiling.
The big job boards are visible, fast, and familiar. Those qualities make them feel comprehensive, but familiarity is not the same as coverage. Companies post directly to their own career pages through ATS platforms like Greenhouse, Ashby, Workable, SmartRecruiters, Recruitee, Personio, Lever, and Teamtailor, and a large share of those listings never get syndicated to the major aggregators at all.
The gap is not a minor rounding difference. It is a separate inventory, and engineers who skip it are searching a subset of what is actually open. The rest of this article breaks down where that inventory sits, how it is distributed across platforms and job functions, and what a search strategy looks like when it accounts for both markets.
The Reference Number: 26,037 Engineering Listings Across Career Pages
Engineering is the single largest department in the dataset, with 26,037 active listings posted directly on company career pages through supported ATS platforms. Of those, 7,073 are marked remote, which is the highest absolute count of remote openings of any department in the dataset. These numbers come from employers publishing through Greenhouse, Ashby, Workable, SmartRecruiters, Recruitee, Personio, Lever, and Teamtailor, collected via public ATS APIs, not pulled from aggregator feeds.
The 26,037 figure is the baseline a software engineer should use to calibrate any search tool. If a platform returns a fraction of that count for engineering roles, the scope is too narrow or the index is incomplete. If it returns several times more, the platform is likely mixing in stale listings, duplicates, or roles that have already closed.
Within that total, Software Engineering as a distinct sub-category carries 1,019 listings from 128 employers. Of those, 214 are marked remote. That sub-category is narrower by definition, because many employers label the same type of role under the broader Engineering department rather than a dedicated Software Engineering tag. The 1,019 number reflects direct-to-career-page postings only, with no aggregator duplication.
The remote count of 7,073 across Engineering is worth noting separately. It is not spread evenly across employers or platforms, and the next sections break down where that concentration sits. Knowing the total first makes it possible to judge whether any individual platform or search is returning a representative slice or a thin one.
How to Read These Numbers: Good, Average, and Thin Coverage
Coverage quality on any engineering job search tool comes down to how many ATS platforms it draws from and how many distinct employers those platforms connect to. Ashby links to 2,713 employers, Workable to 2,675, and Greenhouse to 2,420. A tool that pulls from all three already reaches a substantial share of the direct-from-employer market. Everything below that threshold is, by definition, a partial view.
Platform breadth matters because each ATS attracts a different mix of employers. Recruitee adds 1,812 employers to the picture, and Personio adds 1,275. Both skew toward lower remote density: Recruitee carries 3,356 remote listings out of 28,407 total, and Personio carries 832 remote listings out of 14,454 total. If remote work is a hard requirement, those two platforms alone will not carry the weight, but they still represent employers that post nowhere else.
At the thin end of the range, Lever covers 17 employers with 829 total listings, and Teamtailor covers 2 employers with 32 listings. Neither number justifies treating either platform as a primary source. Their value is specific: a role that appears only on Lever or only on Teamtailor will not show up in any search that skips them, and that gap can matter when the employer in question is one you are actively targeting.
The practical benchmark for good coverage is a tool that connects to at least the top three platforms by employer count, Ashby, Workable, and Greenhouse, while also indexing Recruitee and Personio for the employer segments those platforms serve. Anything drawing from fewer than three of these sources should be treated as a starting point, not a complete search.
Where Remote Engineering Listings Actually Concentrate
Ashby carries 25,087 remote listings out of 44,819 total, the highest remote count relative to total volume of any platform in this dataset. Greenhouse follows with 15,724 remote listings out of 93,817 total, and Workable adds 15,346 remote listings out of 67,567 total. Engineers who treat remote work as a hard filter should build their search around these three platforms first.
The gap between platforms becomes clear when you compare raw remote volume. Ashby's 25,087 remote listings come from a total pool of 44,819, which means remote roles make up a substantial portion of what Ashby-connected employers post. Greenhouse and Workable each carry more total listings than Ashby, but their remote counts are lower. Skipping any one of these three means leaving a significant slice of remote-friendly postings unsearched.
SmartRecruiters shows 9,556 remote listings out of 158,523 total. That raw listing count is the largest of any single platform in the dataset, but the remote volume trails all three of the platforms above. One important note: the listing count for SmartRecruiters is understated for 16 employers due to a collection ceiling of 2,000 listings per employer, so the actual total is higher than 158,523, though the employer count is not affected.
For engineers who need remote work, the platform mix matters as much as total volume. A search limited to SmartRecruiters-connected employers, while skipping Ashby and Greenhouse, misses 25,087 and 15,724 remote listings respectively. Those are not marginal numbers, and they do not overlap with each other, because each platform connects to a distinct set of employers.
What the Big Boards Are Missing and Why It Matters
LinkedIn, Indeed, Glassdoor, Dice, and similar aggregators are where most engineers spend their search time, but none of them are sources for the direct-from-employer market described in this guide. The 70,036 remote listings indexed across all eight ATS platforms in this dataset do not appear as a unified, searchable set on any single major job board. An engineer relying only on aggregators is searching a filtered and delayed version of the market.
The gap is structural, not incidental. Aggregators pull listings after they are posted, often with delays, and they mix employer-direct roles with listings from staffing agencies and third-party recruiters. Listings collected via public ATS APIs reflect what the employer posted on its own career page, with no intermediary in the chain.
The employer count makes the scale concrete. This dataset covers 11,942 employers publishing through supported ATS platforms, and that pool has no direct equivalent on any major job board. Boards like LinkedIn and Indeed each index some of these employers, but not as a clean, filterable group, and not without the agency and recruiter listings mixed in.
Platforms like Arc, Hired, and Otta serve specific niches and are worth using alongside direct-source tools. They are not substitutes, though, because their coverage decisions are independent of what employers post on their own career pages. The only way to search the direct-source market at scale is to query the ATS platforms where employers publish first.
If Your Search Is Returning Thin Results: What to Check
If your engineering search is coming back thin, the most likely cause is platform scope. The market has real volume across thousands of employers, and a tool that skips major ATS platforms will return a fraction of what is actually posted. Checking which platforms your search tool queries is the fastest way to diagnose the problem.
Start with the three platforms that connect to the largest employer pools. Ashby, Workable, and Greenhouse each reach thousands of distinct employers, and their remote-friendly listings are substantial. If your current tool does not pull from all three, that gap alone explains most thin results.
Remote work availability is unevenly distributed across platforms, so filtering for remote early in a search can make coverage problems look like a supply problem. Run the same search without the remote filter first to see the actual listing volume, then apply the filter to assess what is truly available.
Engineers who are moving toward product or growth functions should also expand their department scope. The Marketing department carries 7,131 listings from 1,783 employers, covering roles that frequently require technical backgrounds. Product shows 3,220 listings from 1,232 employers, a pool that includes product engineering, growth engineering, and technical product management titles that would not appear under a pure Engineering filter.
If Your Search Is Already Broad: How to Narrow Without Losing Depth
When a search spans all supported ATS platforms, the listing volume becomes large enough that filtering by department, remote status, and employer size matters more than adding or removing platforms. The right narrowing strategy preserves depth while cutting noise, and it starts with understanding how department labels overlap in employer taxonomies.
The Information Technology department carries 6,958 listings from 398 employers, and the Technology department adds another 2,363 listings from 365 employers. Both categories overlap with Engineering in how some companies tag their roles, which means filtering by department label can split a single job family across three buckets. Filtering by job title instead of department produces cleaner results and avoids counting the same role type twice.
Engineers who are applying broadly but not converting to interviews should audit which ATS platforms their target employers use. A company publishing through Lever sits inside a pool of 17 employers in the dataset, which signals a smaller or earlier-stage organization. That context changes what to expect from the interview process and the offer structure, and it is worth knowing before preparing for a loop.
The Customer Success department shows 1,462 listings from 485 employers, with 743 of those listed as remote. Engineers with client-facing or integration experience who are open to hybrid roles should treat this category as a secondary search layer rather than ignoring it entirely. The remote density here is high enough that a targeted search within Customer Success can surface roles that would not appear under any Engineering filter.
Where This Data Comes From and What It Does Not Cover
This guide is published by JobFerret, and the numbers cited throughout come from the JobFerret index, which collects listings via public ATS APIs from eight platforms: Greenhouse, Lever, Ashby, Workable, SmartRecruiters, Recruitee, Personio, and Teamtailor. Only employers that publish through one of these eight platforms appear in the dataset. Employers that post exclusively on LinkedIn, Indeed, or custom career-page systems outside these platforms are not included.
The department-level breakdown referenced in earlier sections shows 215,214 total listings and 37,585 remote listings, but those figures describe the top 40 rows of a dataset that contains 7,839 distinct department categories. Engineers filtering by department label should treat those figures as a sample of the full distribution, not a complete count of every category in the index.
The collection method pulls from public ATS APIs, which means the data reflects what each platform exposes through its public endpoints. Employers that keep listings behind a login wall or that use private ATS configurations are outside the scope of this dataset.
One structural limitation applies to SmartRecruiters: a collection ceiling of 2,000 listings per employer means the total listing count is understated for a subset of employers on that platform. The employer count across the dataset is not affected by this ceiling, because the ceiling cuts listings per company, not companies themselves.
This dataset covers a large and measurable portion of the direct-from-employer market. It does not represent the entire market, and no claim here should be read as a total count of all open engineering roles globally.