How RemoteDevWork works
A transparent look at where the data comes from, how each listing is read and graded, and what you get as a developer or a hiring team.
Where the data comes from
We read the public job feeds that companies publish themselves through their applicant-tracking systems - the same structured feeds companies use to syndicate their roles to job boards. We do not scrape logged-in networks, and we link every listing back to the employer's own application page.
Each company is identified by its public board handle. We crawl a rotating slice on a schedule, politely rate-limited with a contactable user agent.
The freshness ledger
When we first see a posting we write that date once and never change it. Combined with the date the employer claims, this lets us grade every listing:
- Fresh Younger than most postings like it. Apply now.
- Normal Ageing in line with similar postings.
- Stale An outlier survivor. Often hard to fill, which can mean thin competition.
- Re-stamped Claims to be newer than the day our crawler first saw it.
- Ghost Open for over a year. Probably not a live role.
How we read a listing
Everything is derived with transparent, deterministic rules - no AI language models are involved, so results are consistent and explainable.
We combine the employer's own workplace field, the location text, and the body to decide remote, hybrid or on-site. Only confirmed remote and hybrid roles make it onto the board.
We detect languages, frameworks, databases and tools anywhere in the title or body, then pick the dominant family and list every skill we found.
From the title and opening paragraphs: intern, junior, mid, senior, lead, principal or manager.
A proxy built from skill rarity, how long the role has been open, reposts, hiring volume and application friction. It is an estimate, never a measurement.
How we decide a role is “anywhere”
“Remote” often hides a geography requirement. We only mark a role as Remote - anywhere when the listing says so explicitly (anywhere, worldwide, globally, any location) and we find no restriction that contradicts it.
Signals that disqualify “anywhere” include phrases like “must be authorized to work in…”, “must reside in…”, “only considering candidates in…”, “no visa sponsorship”, or a named country/region. When a constraint is present we surface the real scope instead - region (e.g. EMEA), country only (e.g. US), or unspecified.
What it looks like
Every application you track flows into a live picture of where you stand - and where roles drop off. Here's an example.
Manage the search on a board
Tracking works like a simple kanban board. Each stage is a column, every application is a card, and moving a card updates your funnel straight away. Roles you applied to elsewhere join the same board, so the whole search lives in one place.
Drag a card between columns to change its status. Open a card to add notes as the process moves.
For developers
- Search by skill, stack family, seniority, freshness and competition, plus a true “remote anywhere” filter.
- Save roles and track applications on a drag-and-drop board, including listings you applied to elsewhere.
- A dashboard application funnel, a personal to-do list, and company outcome stats from the community.
- Saved searches with a daily digest, and instant alerts for supporters.
- RSS for anything.
For recruiters & agencies
- API access to search roles, pull daily changes and read company outcome stats - built for sourcing tools and reporting.
- Freshness you can trust: see what is genuinely still open, not what a board wants you to believe.
- Featured listings are coming for employers who want visibility (clearly labelled, never disguised as organic).
- One supporter plan covers both developers and agencies.
Straight talk
Coverage is the bottleneck of any aggregator: we see what companies publish publicly, and no source is perfect. Classifications (especially competition and “anywhere”) are best-effort estimates. We show our reasoning on every listing so you can judge for yourself, and we always link to the original posting.