N S W E

Clarity reduces noise.
Clarity increases impact.

Could I replicate the mind of a hiring leader?

Hiring Compass externalizes what a hiring leader actually wants — before the first resume lands. A living calibration system, not a static rubric.

Clarity reduces noise. Clarity increases impact.

01The Origin

It started as a translation problem.

The earliest spark was a Sourcing Brief Translator, built for Brand Studio — a way to turn a hiring leader's preferences into language a sourcer could actually act on. That small tool kept pulling at a bigger thread: what if the translation didn't have to happen by hand, every time, for every search?

02The Friction

Nearly two years of work surfaced six compounding problems.

High volume with no time to assess. Calibration inconsistency across PEs. Applicant-vs-source duplication. Cross-functional misalignment. Bottlenecks on PE availability. No institutional memory. Hiring Compass was designed to solve all six at once — before there was any performance proof it would work.

High volume, no time to assess quality at speed
Calibration inconsistency across PEs
Applicant vs. sourced duplication of effort
Cross-functional misalignment on what "strong" means
Searches bottlenecked on a single PE's availability
No institutional memory when a role reopens
03The Capability Explorer

Six capabilities. One compounding engine.

Each capability maps directly to one of the six problems above. Click any node to see what it does, why it exists, and the command that activates it.

01
PE Profile Building
Living institutional memory
A living profile per hiring leader — preferences, hard filters, and calibration history extracted from every round and debrief.
"What does this hiring leader actually want?"
02
Candidate Scoring
Fast triage & deep analysis
Yes/maybe/no on resume dumps or sourcing lists — scored against the actual profile, not just the JD. One lens for applicants and sourced candidates.
"Score these candidates"
03
Calibration Outputs
Ready for review
Shortlist with reasoning, detailed breakdowns, and a full calibration sheet ready for hiring-partner feedback — in minutes, not hours.
"Build a calibration sheet"
04
Rubric & Scoring
Shared language for "good"
Role-specific rubrics built from the JD, spot doc, and profile. Hard filters run first — no score assigned to candidates who don't clear.
"Update the profile with this feedback"
05
Cross-Role Routing
Catch people mid-funnel
Flags candidates declined from one role who may fit another active opening. Active roles only — closed roles stay invisible to routing.
"Cross-role routing check"
06
Sourcing Direction
Get up to speed fast
Targeting language derived from the profile — for coverage gaps, sourcing jams, or picking up a search you've never worked before.
"Sourcing brief for this role"

PE Profile Building — Living Institutional Memory

Every hiring leader gets a two-file profile: preferences, hard filters, and calibration history — plus a scoring skill that knows how to evaluate candidates against that history. When the hiring leader is unavailable, their judgment isn't lost. When a role reopens, the profile is already seeded.

04See It Work

From calibration to recommendation.

This is what actually happens between a candidate landing in the pipeline and a recommendation reaching a hiring partner.

Input
  • Role context
  • Hiring-leader calibration history
  • Historical signal from past rounds
  • Candidate evidence (resume, portfolio, notes)
Hiring Compass

Hard filters → dimensions → history

  • Hard filters run first, no exceptions
  • Scored across weighted evaluation dimensions
  • Checked against revealed preferences
  • Calibrated against past rounds for this role
Output
  • Recommendation (Strong Yes / Maybe / No)
  • Evidence supporting the call
  • Watch-outs to raise in review
  • Routing opportunity + next action
05Representative Output

The moment it stops being conceptual.

This is a sanitized reconstruction of an actual Hiring Compass scorecard's architecture — every candidate scored returns a structured recommendation in this format, with the disclaimer already built in.

Hiring Compass · Candidate Scorecard

Sanitized reconstruction — role and identifiers generalized

2026
Invite Partner: Bri Majors
Candidate Ref. 24-118
Senior Brand Designer
16
Strong Yes
Advance to hiring leader
RoleWeb Experiences
LevelL5
Hard Filters✅ All cleared
Threshold14+ = Strong Yes
Key Takeaway

Strong Yes. Systems-first orientation with clean modern brand aesthetics — exactly the signals this hiring leader has marked as top priority. Portfolio shows sustained depth across brand systems and scalability, not just execution polish. Probe AI fluency in production before advancing to screen.

Dimension Scores
D1
Interactive / Web Craft
Portfolio leads with systems and scalability — strong web-native sensibility, not print-first.
3
D2
Brand Expression / POV
Clean, modern aesthetic with clear point of view. Considered and restrained, not trend-chasing.
3
D3
Design Systems / CMS
Systems focus is the lead signal in the profile. Scalability framing suggests systems-level thinking.
3
D4
Motion / Interactive Range
Portfolio shows interactive range but motion depth is lighter — not a dealbreaker for this role.
2
D5
AI Fluency (in production)
Not confirmed from available materials — probe required. Tool use in actual workflow, not ideation.
2
D6
Taste / Restraint / POV
Aesthetic signals are strong — modern, clean, restrained. Aligns with the hiring leader's confirmed taste signals.
3
Total ScoreStrong Yes threshold: 14 · Max: 18
16 / 18
Recruiter Dig-Ins Before Advancing
AI fluency in production — which tools are in your actual workflow? Ideation use only is not sufficient for this role.
Portfolio depth on the systems side — walk through how a design system was built or scaled. Looking for process and tradeoffs, not just the output.
CMS experience — has this shown up in shipped work, or primarily in brand/identity contexts?
Sanitized reconstruction of an actual Hiring Compass scorecard. Candidate identity, hiring-leader identity, req ID, and portfolio/LinkedIn URLs have been removed or generalized. This is a starting point for conversation, not a final hiring determination — human judgment is always determinative.
05bSystem Evolution

The scorecard was the beginning, not the ceiling.

Hiring Compass didn't stop at scoring individual candidates. The same evaluation logic became the foundation for something bigger — a way to centralize what an organization knows about its own hiring.

Calibration EngineScoring individual candidates against revealed preferences
Structured IntelligenceIndividual scores compound into durable, queryable signal
Hiring HubA centralized home for roles, candidates, and calibration in one place
Institutional MemoryNothing starts from zero — the organization's hiring judgment persists
06The Learning Loop

Every search improves the next search.

Calibration isn't a one-time setup. It's a loop that runs every time a hiring leader reacts to an output.

Calibration
Signal
Profile Evolves
Next Evaluation
Closed-Search Learning
back into the system
07The Transformation

How the operating model changed.

BeforeAfter
Calibration work got recreated from scratch on every search, even for hiring leaders worked with before.
Institutional knowledge compounds automatically — every calibration round sharpens the next.
Preferences lived in Slack threads and memory. When the hiring leader was unavailable, the search stopped.
Documented, calibrated preferences carry forward. Searches move even when the hiring leader is unavailable.
Resume review was inconsistent — different bars depending on who was in the room.
Review is faster and consistent — scored against actual calibration history, not just the JD.
Applicants and sourced candidates were judged through different lenses — same person, different outcome.
One rubric, one lens — applicant or sourced, same criteria, no duplicated effort.
Closed searches lost their value. Every reopened role started from zero.
Every search improves the next one. Closed roles become calibration baselines.
AI was a productivity tool — faster drafts. The strategic judgment still sat entirely on the recruiter.
AI reduces cognitive load without replacing judgment — human review stays determinative.
08Designed for Adoption

Built to onboard, not just to work.

Bri didn't just build functionality — the system shipped with onboarding, documentation, commands, and a live demonstration model so any Invite Partner could pick it up.

/hiring-compassActivate inside Claude
Choose RoleInvite Partner or hiring PE
Choose TaskScore, calibrate, route, source
Run SystemStructured recommendation returns
Feedback LoopImproves the next output
09Impact, Observed

700+ candidates have moved through the early system.

In observed use, a 25-resume review that once took roughly 60 minutes now takes about 7–8 minutes — around 87% less review time, or 7.5–8.5x more throughput.

These are observed working comparisons from real searches, not controlled scientific benchmarks.

700+
Candidates Processed
~60 → 7–8
Minutes per 25-Resume Review
~87%
Less Review Time
7.5–8.5x
More Throughput
Corroboration

Building Claude skills for application reviews with Bri helped us create a shared brain so much faster. We were able to externalize and align on our candidate criteria, and get calibrated on the role with lightning speed. It made the application review process so much smoother. The shared skills also helped me avoid bias and inconsistency in reviewing applications—it reminded me of the skills and experience we'd previously decided was most important for the role.

— Alaine MackenzieVP of Product Design, CoreX AI · Gusto, at time of feedback
10Backstage

What's shown. What stays backstage.

Shown on this page
  • The six capabilities and what each one does
  • The calibration → recommendation flow, end to end
  • A representative output's anatomy and format
  • How the system learns across searches
  • Before → after operating change
  • The adoption model — how a new PE gets onboarded
Stays backstage
  • Exact prompts and scoring weights
  • Private PE profiles and calibration history
  • Real candidate identities and req IDs
  • Internal Gusto implementation details

Hiring Compass eventually changed how I think about recruiting altogether, and helped inspire brimajors.studio itself.

Alignment suggestions only, not final hiring determinations. Human review is determinative.