AI matching

A ranked shortlist the moment the brief lands. With the receipts.

Not a black-box percentage. Kepler turns the client brief into a scorecard, scores everyone in your database against it, and gives every verdict the line from the CV, call or email that earned it.

Start free trialSee how it works
Senior Data EngineerNorthwind Capital · London · Open
Criteria · 4 Re-run
Example matches for a Senior Data Engineer brief.
CandidateMatchRole titleCurrent employerSkills
Sophia Bennett92StrongSenior Data EngineerMonarch AnalyticsPython, Spark, AWS, Terraform, dbt
Amir Hassan81GoodData EngineerNorthwind CapitalPython, Spark, EMR, SQL, Airflow
James Okafor74GoodSenior Data EngineerJuniper SystemsSpark, Azure, Databricks, Kafka, Scala
Mei Tan68WeakEngineering Manager, DataAtlas CloudAWS, Leadership, Python, Snowflake
Elena Rossi55WeakAnalytics EngineerFormadbt, SQL, Python, Looker
Oliver Chen49Off specPlatform EngineerNorthwind CapitalAWS, Kubernetes, Terraform, Go
01

Every number traces to evidence

Unpack a score and you are reading criterion verdicts, each sourced to the document that earned it.

02

Clients get the reasoning

The shortlist you send carries the why, so it is defensible in the first client call rather than after it.

03

It admits ignorance

Nothing in the record for a requirement? The verdict is “Not stated”. It never guesses to fill a gap.

Under the hood

The brief becomes a scorecard. The scorecard ranks the room.

Kepler reads the CVs, notes and conversations in the eligible pool — not just the profile fields — and ranks the strongest first. Open any row for the assessment, a verdict on every requirement, and the line each one rests on.

See it on your own data
Senior Data EngineerNorthwind Capital · London · Open
Criteria · 4 Re-run
Example matches for a Senior Data Engineer brief. Select a row to read its evidence.
CandidateMatchRole titleCurrent employerSkills
Sophia Bennett92StrongSenior Data EngineerMonarch AnalyticsPython, Spark, AWS, Terraform, dbt
Amir Hassan81GoodData EngineerNorthwind CapitalPython, Spark, EMR, SQL, Airflow
James Okafor74GoodSenior Data EngineerJuniper SystemsSpark, Azure, Databricks, Kafka, Scala
Mei Tan68WeakEngineering Manager, DataAtlas CloudAWS, Leadership, Python, Snowflake
Elena Rossi55WeakAnalytics EngineerFormadbt, SQL, Python, Looker
Oliver Chen49Off specPlatform EngineerNorthwind CapitalAWS, Kubernetes, Terraform, Go

Select a candidate to read the evidence.

Sophia Bennett

1 of 6

92

judge score

Strong match

3 of 4 criteria met

Assessment

A strong fit for the brief. Sophia has run a production Spark platform on AWS and has managed a team through a migration. Financial services exposure is via a payments client rather than a bank.

Candidate

EmployerMonarch AnalyticsLocationLondonStatusActiveLatest degreeMSc Computer ScienceEmailsophia.bennett@monarch.exampleProfileSophia Bennett

Criteria · 3 of 4

  • Financial services“Two years on a payments client; no bank or fund experience stated.”
  • Python & Spark in productionmust“Built the ingestion platform in Python and Spark, processing 40 TB a day.”
  • AWS data platform“Ran the platform on AWS: Glue, Redshift and S3, with Terraform.”
  • Led a team“Managed four engineers across two squads through the warehouse migration.”

↑ ↓ next candidate · Esc close

You stay the recruiter

The ranking arrives done. It never arrives final.

Hard limits run before scoring so the pool is one you could actually place. After it, your pins override the order and the score keeps explaining itself underneath.

Hard limits before scoring

£90k+London · hybridRight to work: UK

Applied to the pool first, so nobody is scored who could never take the job.

Your order overrules the score

  1. 1Sophia Bennett92Strong
  2. 2James Okafor74Good pinned by you · was #3
  3. 3Amir Hassan81Good
  4. 4Mei Tan68Weak
  5. 5Elena Rossi55Weak

The pin sticks and the reasoning stays attached. The score explains; you decide.

What you are looking at

Three parts, and none of them a black box.

The scorecard you tune, the verdicts it produces, and the inbound applicants that arrive already scored against the same rubric.

Senior Data Engineer brief pasted

  • Python & Spark in productionmust
  • AWS data platformhigh
  • Led a teamhigh
  • Financial servicesmedium
  • London or hybridmust

Extracted from the brief — tune anything before a CV is read.

The spec becomes a scorecard

Paste the brief and the criteria and weights are pulled out for you to tune — musts, highs and mediums, all editable.

Sophia Bennett Strong

  • Python & Spark in production“Built the ingestion platform in Python and Spark, processing 40 TB a day.”
  • AWS data platform“Ran the platform on AWS: Glue, Redshift and S3, with Terraform.”
  • Led a team“Managed four engineers across two squads through the warehouse migration.”
  • Financial services“Two years on a payments client; no bank or fund experience stated.”

Scored against everything you know

The judge reads CVs, notes and conversations, and every verdict cites the exact source it came from.

Applications 2 new · via job board

  • Iris ColeApplied Sat 09:12 · CV parsedDeduped91Strong
  • Jonah ReyesApplied Fri 17:48 · CV parsedNew74Good

Inbound, scored on arrival

Applicants from your hosted job board land parsed, deduplicated and scored against the same rubric as the rest of the pool.

With receipts

Scores you can say out loud.

A match score is a claim about a person. Kepler attaches the proof, so the number survives contact with your client.

  • Sourced, criterion by criterion. Met, partial, missing or not stated — each with the sentence from the profile behind it.
  • Filtered before it is scored. Salary, location and right to work apply to the pool first, not as a caveat afterwards.
  • Measured before it ships. Matching runs under a standing evaluation program; recall and precision gates clear before a scoring change goes out.

Questions, answered.

Can a recruitment CRM match candidates already in my database to a new job description?

Kepler does. Paste the client’s brief and Kepler pulls the criteria and weights out of it as a scorecard — musts, highs and mediums — which you tune before a single CV is read. It then scores everyone in your database against that scorecard and ranks the strongest first.

Can I see why someone scored what they scored?

Yes, and that is the point. Open any score and you are reading criterion-by-criterion verdicts, each one sourced to the CV, call or email that earned it, so the shortlist carries its reasoning into the first client call.

What happens when a record says nothing about a criterion?

The verdict is “Not stated”. Kepler does not infer a requirement from an adjacent skill or guess to fill a gap — a silent record stays silent, and you can see exactly which requirements still need a conversation.

Can I overrule the ranking?

Always. Pin someone higher or move them down and your order sticks. The score stays visible underneath with its evidence attached, so a colleague can see both what the judge thought and what you decided.

Does matching read candidates I am not allowed to see?

No. Scoring runs over the records your permissions already allow you to open, inside your workspace, and the shortlist you get back is bounded by the same rules as the rest of the app.

Better together.

AI assistant

Ask your database a question and get an answer you can act on.

CV parsing & migration

Bulk CVs and old-system exports in, searchable profiles out.

Job board

Publish roles on your own domain and get applicants parsed and scored.

Start your free trial.

We set up your workspace, migrate your data and walk you through it.

  • Free migration from your current system
  • No card, no minimum seats
  • A reply within one working day

Prefer to talk first? Book a demo · Privacy policy