Product
A search engine for the resumes you already own
Most recruiting teams sit on years of resumes they can’t really search. Moneko reads every one of them, understands what the candidate has actually done, and lets you query the whole archive the way you would brief a colleague: by pasting the vacancy.
What happens between “upload” and “shortlist”
Every step is automatic and runs per dataspace. You only see the results — but here is what the results are made of.
Ingest
PDF, DOC and DOCX files are uploaded straight to secure storage; Word files are converted to PDF so every resume opens in the viewer. Text can also arrive through the API.
Parse
In parallel, AI extracts title, name, location and typed contact details; builds up to 128 canonical keywords with synonyms; writes a concise summary and a one-line brief; and rewrites the resume into a clean text used for embedding.
Index
Each candidate gets a 3072-dimension embedding for semantic search plus a full-text index over title, name, location, contacts, custom fields and keywords. Changed files re-trigger parsing automatically.
Search & rank
Your vacancy is embedded and matched against the vector index (optionally within a tag). The top 100 are then compared head-to-head by AI in a tournament-style pass, so the order reflects the whole query — not just similarity.
Assess
As you scroll, each visible candidate is assessed against every requirement — Good, Fair or Poor — with a sentence of evidence. Results are cached per resume version.
Built around how recruiters actually work
Each capability below is in the app today.
Search by pasting the vacancy
No boolean strings, no filters to configure. Paste the whole job description, or describe the person you need in plain language.
AI distills the query into a ranked list of concrete requirements — technical skills, certificates, licences and other must-haves — and ignores noise like salary, schedule or generic soft skills.
Long queries get a short summary and a search title, so the history stays readable.
Ctrl/⌘ + Enter to run a search from the keyboard.
Factor-by-factor match assessment
Every candidate in the results is scored Good / Fair / Poor against each requirement — with a one-sentence justification grounded in their resume.
Assessments are evidence-based: a skill merely listed without supporting experience is capped at “Fair”, outdated or brief exposure is downgraded, and aspirations are not counted as experience.
Hover a badge to read why. Assessments are cached per resume version and re-computed when a candidate’s files change.
Candidates without a resume, or with files that are not actually resumes, are flagged instead of guessed.
Semantic and keyword search, together
Vector search finds people who match the meaning of the vacancy; keyword search with synonyms, highlights and did-you-mean finds the exact term you typed.
AI search: the query is embedded and matched against a rewritten, embedded version of every resume; the closest 100 are then compared head-to-head by AI to produce the final ranking.
Instant search: type a skill, a name, a phone number or a location — results highlight the match in the title, location, contacts and extracted keywords (including synonyms).
Autocomplete suggestions and spelling corrections come from your own database, not a generic dictionary.
Automatic resume parsing
Drop in PDF, DOC and DOCX files. AI extracts the job title, name, location, contact details, keywords and a two-sentence summary — no templates required.
Word documents are converted to PDF automatically, so every resume opens in the built-in viewer.
Contacts are recognised by type: phone, email, LinkedIn, website, WhatsApp, Telegram and more — and shown as clickable fields on the candidate page.
Up to 128 canonical keywords plus synonyms per resume power the instant search.
Each candidate can hold several files (different resume versions, cover letters); the newest version wins when they disagree.
A candidate database you control
Browse, tag and bulk-edit candidates. Filter any search by tag. Every field is editable, and the change log shows who changed what — including the AI.
Tag candidates by pipeline, source, seniority or anything else; add or remove tags for many candidates at once.
Run an AI search within a single tag to shortlist from a specific pool.
A per-candidate log records creation, field changes, attached or deleted files and tag changes, with the author — a human or the parser.
Curate the shortlist
Pin the people you want to keep, hide the ones you don’t, and load more when you need a bigger pool. The search remembers your decisions.
“Load more” pulls the next batch of candidates, excluding everyone already shown, pinned or hidden.
Add a candidate to a search manually when you already know who belongs there.
Results are stored per search, so you can come back tomorrow and continue where you left off.
Built for teams
A dataspace is your team’s shared workspace. Invite colleagues, see each other’s recent searches, and keep one candidate database instead of ten spreadsheets.
Recent searches are split into yours and your team’s, right in the sidebar.
Admins manage members, roles and API keys for the dataspace.
One account can belong to several dataspaces — handy for agencies working with multiple clients.
Search history that is actually useful
Every search keeps its query, requirements, results and your pins. Rename it, revisit it, or delete it when the role is filled.
Filter history by author and date.
Searches are also indexed, so you can find an old search by what it was about.
REST API and API keys
Push candidates from your ATS or CRM with a single endpoint. Moneko parses and indexes them just like uploaded files.
POST up to 100 candidates per request with your own IDs; re-sending unchanged text is a no-op, changed text triggers re-parsing.
Create, disable and delete keys per dataspace. Each key acts under its own name, so the change log stays attributable.
Privacy by design
Your resumes stay yours. Files and search indices are hosted in the EU; AI providers process data under API terms that exclude training on your content.
Resume files and the search index are stored in AWS and Elastic Cloud in Frankfurt (eu-central-1).
AI requests are made without storage on the provider side (OpenAI Responses API with store disabled; Google Gemini API for embeddings).
Strict tenant isolation: every record is scoped to your dataspace, and access is verified on every request.
Your data, your dataspace
Resumes are personal data. Moneko is built so that you stay in control of them.
EU hosting
Resume files and search indices live in Frankfurt (AWS and Elastic Cloud eu-central-1).
No training on your data
AI requests go to OpenAI and Google APIs with storage disabled; providers do not train on API content.
Tenant isolation
Every record and index is scoped to your dataspace; membership is checked on each request.
Audit trail
Who uploaded, edited, tagged or deleted what — people and API keys alike — is logged per candidate.
Modern auth
Sign-in, organisations and invitations are handled by Clerk; dataspace admins manage members and keys.
Details in the Privacy Policy.
Bring your resumes. Moneko does the reading.
Start with a free dataspace, upload a batch of resumes and run your first AI search in minutes. Questions first? We answer email personally.
hi@moneko.ai