WeRemoteIT WeRemoteIT
AI Matching · Synapse
Explainable scores · Remote IT

Match every application
to the role — not the résumé keywords.

WeRemoteIT scores candidates against verified job requirements so clients and talent spend less time on mismatches and more time on conversations that can close.

0–100

Composite match score

Title fit, experience depth, and skills coverage — with a readable reason on every row.

3

Core signal weights

Title 45% · Experience 25% · Skills 30%, plus domain guardrails and remote fit bonuses.

Live

Inbox & bots

Scores appear in the candidates pipeline, talent job fit, and chat-native apply flows.

WeRemoteIT AI Matching1 / 5
WeRemoteIT WeRemoteIT
Deterministic engine
How the score is built

Three weighted signals. Domain guardrails. Explainable output.

Job title similarity

45%

Role-family mappings (e.g. DevOps ↔ SRE / Platform) — perfect, closer, or weak adjacency — not raw string equality.

Years of experience

25%

Compares required seniority to the talent profile; bonuses for meeting the bar, penalties for large gaps.

Skills alignment

30%

Required skills vs verified profile skills — coverage, missing skills, and stack depth signals.

Cross-domain mismatches (e.g. healthcare title vs engineering role) take a hard penalty so keyword coincidence does not inflate the score. Remote and education signals can adjust the final number.
WeRemoteIT AI Matching2 / 5
WeRemoteIT WeRemoteIT
Bands & surfaces
What teams see

Score bands in the pipeline — sorted highest first.

Perfect / Strong

High title + skills fit. Prioritize for interview.

Good / Potential

Solid adjacent fit. Worth a shortlist review.

Fair

Partial coverage. Use filters and reason tooltips.

Weak

Low alignment or domain mismatch. Deprioritize.

Client surfaces

  • Candidates inbox grouped by job
  • Progress bar + match % + reason tooltip
  • Manual “Analyze Match” recalculation
  • Filter by score threshold and vetting tier

Talent & channels

  • Job fit score on talent job board
  • Auto-score on apply (DB trigger)
  • Telegram / Discord / Slack apply parity
  • Same identity, same score math everywhere
WeRemoteIT AI Matching3 / 5
WeRemoteIT WeRemoteIT
AI assists vs score
Two layers

Generative AI helps write. The match engine decides fit.

Deterministic match engine

The 0–100 score is rule-based and repeatable: title mappings, experience math, skills coverage, domain guards.

  • Explainable breakdown and reasons
  • Runs automatically on apply
  • Bias-aware: skills and role fit first

AI assists (job / CV)

LLM helpers improve inputs so matching has better signal — they do not replace the score formula.

  • AI job description / skills fill
  • CV / profile parse into structured fields
  • Better JD → higher-quality matches
Pipeline view for buyers: Profile parse → Skill mapping → Role scoring → Human review with explainable scores on web and chat.
WeRemoteIT AI Matching4 / 5
WeRemoteIT WeRemoteIT
Next step
See it live

Review scored applicants — or browse roles with fit labels.

Matching is part of the chat-native ATS: post once, score everywhere, message the shortlist without leaving the thread.

Protocol docs

https://weremoteit.com/docs/client/matching.html

Insights

https://weremoteit.com/insights/ai-matching

Sales

projects@weremoteit.com

Read AI matching insight Protocol matching guide Pricing PDF / page
© 2026 WeRemoteIT · Registry WR-7729-IDX5 / 5