What Intelletto actually is
Most résumés don't get read.
We read all of them.
Intelletto plugs into the ATS you already run. It reads every résumé that lands there, scores it against the actual job you posted, and tells you why — with the source passage attached, so a regulator, a hiring manager, or the candidate can check the work.
Measurable lift · current quarter
The numbers move in the direction you'd want.
Four metrics every talent leader is measured on. Bars are drawn to scale — the longer the bar, the bigger the shift.
OK, so what does it actually do
Three things. One underlying engine.
Think of Intelletto as three products sharing a single pipeline and a single audit trail. Most teams switch one on first — usually Human Capital Intelligence — and add the others as the value compounds. None of them require a replatform.
Read everything
Turn 200 messy PDFs into 200 comparable profiles.
Every résumé gets read, structured, normalised against your skills taxonomy, and stored as a profile you can compare, sort and search. The audit trail goes back to the bytes of the source PDF.
Score what matters
Scores that mean something specific.
Nine buckets scored against each role, weighted per seniority band. Then we follow the people you hired and check the score against their 30/90/180-day outcomes — so the rubric tightens after every cohort, not every quarter.
Show your work
Decisions you can defend.
Every recommendation ships with reason codes, a confidence score, and the source passages that justify it. Every sealed scorecard produces a 14-page Audit Report — open it inline, or hand the signed bundle to a regulator who runs our open-source verifier on their own machine.
How it plugs in
We plug in. Nothing else changes.
Two sides of one commitment: no new system for anyone to log into, and no automated decisions until your team has watched the model long enough to trust it.
The sidecar model
We talk to your ATS through APIs. That's the whole integration story.
The change-management slide is one line: a new feed appeared inside the tool your team already uses. No new login, no new dashboard, no "is everyone trained yet" meeting. Intelletto reads from and writes back to the systems below.
The rollout path
Shadow → Gated → Automate. You decide how fast.
Shadow.
Intelletto runs in parallel and produces scores nobody acts on. You compare them to what your team would have done. The "is this thing any good?" phase.
Gated.
Recommendations show up in the recruiter's view, but every one needs a human approve / edit / decline. Every override gets logged and feeds the model.
Automate.
Only on the requisitions you say so, only above the confidence threshold you set. Pull the threshold back any time. There is no one-way door.
Under the hood
How a résumé turns into a score you can defend.
stages
gates
buckets
modifiers
modes
The twelve stages
Twelve stages. Five quality gates. Same path, every time.
The first four stages answer four boring questions — what document, who owns it, have we seen it, what does the page actually say. The middle stages do the AI work. The last stages produce the artefacts a recruiter looks at.
Register & track
A file lands. We assign a tracking ID, hash the bytes, link it to a job code if there is one, and remember exactly when and where it arrived.
Have we seen this person?
Byte-level match catches the obvious. Email + name + phone catches the "same person, slightly different PDF" case. If it's a dupe, the run ends here.
04
Actually read the document
Google Document AI handles the OCR and layout. We strip the headers, footers, watermarks, page numbers — anything that isn't the content the candidate wrote.
Gate A · LosslessPull out the structure
Gemini extracts skills, roles, achievements, certifications, languages and executive signals into a typed JSON schema. Every claim is bound to the passage that supports it — or it doesn't make it through.
Gate B · SchemaSpeak the same language
"K8s", "Kubernetes" and "container orchestration" all collapse to the same node in your taxonomy. Certifications expand into the skill basket they imply.
Gate C · EvidenceLook around the web
LinkedIn, personal site, GitHub. We actually fetch the pages (not pretending to) and corroborate the claims. New skills found here carry the source they came from, forever.
09
Freeze the inputs before scoring
All five gates evaluated, every verdict logged. If the candidate gets through, we snapshot everything that's going to feed the scorer and hash it — so a re-score next year reproduces the same number bit-for-bit, or you know exactly why it didn't.
Gate D · Dedup Gate E · Artefacts11
Score against the actual JD
The nine buckets get weighted for the seniority of the role. Five modifiers refine the base score with evidence quality, outcome signals and two risk rubrics. The full scorecard — every bucket, every modifier, every passage — is saved permanently and never overwritten.
Hand the recruiter something they can use
Ranked shortlist. Per-candidate scorecard with matched + missing skills. Risk flags. Quality checklist. An audit packet a regulator can read end-to-end — all landing inside the tool the recruiter already uses.
Five quality gates
Five questions we ask before a score is allowed to happen.
A score built on broken inputs is worse than no score — it's a number someone will trust. So before the scorer runs, every candidate passes five gates. Two are hard stops. Two are warnings the recruiter sees. One is a quality summary. Whatever happens, every verdict ends up in the audit packet.
Did the OCR actually read every page? Did anything drop silently?
Flag it. A human re-reads. The pipeline continues.
WARNDid the AI return the exact JSON shape we asked for?
Hard stop. A malformed extraction is not allowed to become a score.
BLOCKDoes every claim trace back to a passage in the source document?
Show the low-evidence claims to the recruiter. They decide.
WARNHave we already processed this person in your pool?
End the run as a duplicate. The original record stays untouched.
BLOCKURL quality, bounding boxes, normalisation hits, work history present?
Attach a quality summary. The score stays visible.
PASSWhere the AI actually fires
AI isn't a feature here. It's running in about two dozen places.
Most platforms bolt AI onto one step — usually résumé parsing — and call it done. We use structured-output AI all the way across the hiring loop. Every call is locked to a schema and bound to an evidence contract, so we get back something we can actually use — not just "text the AI made up."
Source & search
Find the right people, faster. Type what you actually want and the search figures out the intent.
Understand the candidate
Read every claim, then go check it. Normalise the skills, fetch the websites they listed, read the shape of the career.
Define the role
The JD itself is AI-shaped. "Ninja Engineer III" resolves to a real occupation; certs carry their skill basket.
Score the fit
A rubric that knows the role. Bucket weights shift by seniority; a three-pass verifier catches paraphrased experience.
Engage & interview
Messages a candidate would actually open. Briefs that surface the gap-and-strength shape before the call.
Police its own bias
A separate auditor reviews everything the platform generated, across four bias dimensions. Never blocks. Always logged.
How we score
Nine buckets. Four seniority profiles.
A graduate engineer and a COO can't share the same rubric. What gets a junior shortlisted is not what gets a chief operating officer shortlisted — so we don't make them. The same nine buckets exist for every role, but the weights flip as the job gets more senior. "Can you do the work?" becomes "can you scale it?" becomes "can you own the function?" becomes "can you stand behind it?"
Nine scoring buckets · emphasis at Junior/Mid
Weights flip as seniority rises
A candidate with AWS Solutions Architect Pro probably knows what EKS is — even if their CV never says it.
The JD asks for "EKS" in those exact three letters; the naive scorer says zero match. We link industry-recognised certifications back to their underlying skill basket through the DoL occupation taxonomy. Hold the credential, carry its skills — at a confidence-weighted match that reflects how recent and how tight the mapping is. Nobody gets penalised for writing a tight CV instead of an exhaustive one.
Modifiers · applied after the base score
The bits the base score can't see on its own.
The base score answers "how well do their skills line up with the requirements?" — a real question, but not the whole one. Five modifiers add to and subtract from the base to catch evidence quality, quantified outcomes, and two flavours of risk. Every one is itemised on the scorecard, so a recruiter sees exactly which modifier moved the score, by how many points, and why.
🎯Data Fusion Confidence
Clean, exact taxonomy matches boost. A lot of fuzzy "probably the same thing" fallbacks pull it back.
🔗Evidence Coverage
What share of the JD's requirements have a matching passage? Above half, a bump; below it, pulled down to match.
📈Outcome Evidence
Did they quantify anything? "Cut infra cost 32%", "grew ARR $14M", "scaled the team 6→40". Cheap, strong trust signal.
🛡️Stability Risk
Seven factors: job-hopping, average tenure, gap pattern, continuity, momentum. Not to punish short stints — to flag when the shape says this won't stick.
🔄Role-Switch Risk
Seven factors again: gap to the target role, seniority alignment, skill recency (decayed over 24 months), moving across a never-lived function.
We police our own AI
Every word the AI writes gets read by a different AI.
Worth being clear: the auditor only looks at what Intelletto wrote — not the candidate's résumé, not the hiring manager's notes. Just the AI-generated text: job descriptions, outreach drafts, score rationale. Spot loaded language, surface the finding, suggest a rewrite. A human always decides. It never blocks the work, and every finding gets logged.
Gendered language
"Rockstar", "ninja", "aggressive", excessive "nurturing" — these quietly tilt the applicant pool. The auditor suggests neutral phrasing.
genderAge signals
"Digital native", "young and energetic", "recent grad". The shortcut is age. The auditor names it and asks what the underlying skill actually was.
ageRace, ethnicity & origin
"Native speaker", "no accent", geography shorthand that's really about where someone's from. Often unintentional. It proposes a fluency requirement instead.
originDisability & family
Physical-ability presumptions ("must be able to…" when the role doesn't require it) and family-status filters that don't predict performance.
disability / familyFour modes, one pipeline
Pick the path that matches how you actually hire.
Same twelve stages, four different paths through them. The mode is a per-requisition — or per-source, or per-team — choice. You can use all four in the same week without anyone's permission.
Read, normalise, gate — don't score yet.
Use when you're building a pool of pre-qualified people for roles you haven't opened yet.
Take a pool candidate and score them against a JD that just opened.
Use when a req opens, you already have great candidates in the pool, and you want answers in minutes.
The whole thing: read, score, shortlist, in one pass.
Use when it's standard inbound for a live req. The most common path.
High-volume ingest, plus a clean Intelletto-format résumé per person.
Use when you're migrating from another ATS, importing an agency dump, or absorbing a historical pool.
The modes are just the start. Skill taxonomy bucket weights auto-score thresholds knockout criteria RBAC permissions culture-fit pillars — everything that matters is configurable per-tenant. We're not trying to impose a hiring philosophy on you. You bring the philosophy. Intelletto executes it.
Why you can trust the number
Five reasons every score holds up when someone asks "why?"
The person being scored, the manager doing the hiring, and the regulator reviewing the process all eventually ask the same question: where did this number come from? A scoring engine that can't answer that shouldn't be making recommendations about people's careers. Here's how Intelletto answers it.
01 · Provenance — every score traces back to the bytes
02 · Evidence per claim
No score without a source passage.
Every bucket exposes its matched and missing skills, with the exact line from the document next to each one. If we can't cite what we're crediting, we don't credit it — that's what the evidence gate is for.
03 · Versioning, not overwrites
Re-scoring never destroys history.
Edit the taxonomy, tweak the JD, adjust a weight — none of it overwrites yesterday's score. It creates a new scorecard_version alongside the old one. The original score, snapshot and rubric all stay intact.
04 · Configurable, not magic
Every weight is yours; every change is logged.
Bucket weights, modifier ranges, gate thresholds, and what counts as "the same skill" are all tenant-owned. Every change writes a record — so a regulator can reconstruct exactly what the rubric looked like on the day a decision was made.
05 · One-click Audit Report
A 14-page report and a signed bundle, on demand.
Every sealed scorecard ships with a cover sheet and verdict, bucket-by-bucket evidence, the full SHA256 chain from PDF to score, the pipeline stage outcomes, and the rescore history. Recruiters open it inline.
Auditors download the signed bundle and run our open-source verifier — on their own machine, against the immutable input snapshot — and get back a single line. Not "trust our signature": re-run the scoring yourself and see you get the same number. That's the bar.
Governance is baked in
Not an afterthought. Not a checklist someone fills in after launch.
Security, privacy, fairness, reliability — the four words every enterprise software page lists in the footer. Worth saying what they actually mean here.
Security
Encryption in transit and at rest. Role-based access with scoped tokens. Least-privilege defaults at every pipeline stage — no service has more read or write power than it needs for its specific job.
Privacy
Aligned with GDPR, PDPA and similar regimes. Data access is purpose-limited — what you can read is gated by why you're looking. Every extraction and decision lands in an audit trail you can export.
Fairness
Fairness monitors run continuously, not on request. Every artefact is signed. Every scoring run is reproducible. A bias report — the kind a regulator asks for — is exportable in one click.
Reliability
p95 latency targets per stage. Structured retries on calls that can flake. Graceful degradation when a downstream service is having a moment. One-click rollback at every stage — because sometimes you just need yesterday's pipeline back.