Full Feature Set

Everything you need.
Nothing you don't.

HireSignal was purpose-built for HR teams. Every feature answers one question: does this engineer have what it takes for this role?

Deterministic scoring — not LLM guessesZero data sent to OpenAI / AnthropicNYC LL144 · EU AI Act · EEOC compliant
Read the full scoring methodology →
Feature 01

11-Dimension Scoring Engine

A deterministic, explainable scoring model across eleven HR-relevant dimensions — each with a per-dimension confidence band. Reproducible every time — no black boxes.

Profile Completeness — name, bio, location, links (max 10 pts)
Repository Quality — descriptions, topics, licensing (max 25 pts)
Community Impact — stars, forks & external OSS activity (max 10 pts)
Contribution Consistency — streak, diversity, cadence (max 20 pts)
Technical Breadth — language & topic diversity (max 15 pts)
Social Proof — followers, public presence (max 5 pts)
Commit Quality — message discipline & atomicity (max 20 pts)
Recency — how live is their activity right now? (max 10 pts)
PR Review Quality — external code review engagement (max 10 pts)
OSS Contributions — external project contributions (max 10 pts)
Documentation Quality — READMEs, descriptions, personal site (max 10 pts)
Feature 02

LLaMA 3.1 AI Analysis

A self-hosted LLaMA 3.1 8B model enriches every report with candidate-specific insights — not generic templates.

Plain-English executive summary for hiring managers
5 custom interview questions from the candidate's actual stack
Red flag detection (star-farming, account age mismatch, gaps)
Standout factor identification
AI-authored hire verdict with override transparency
Zero data sent to third-party LLM providers
Feature 03

Commit Quality Analysis

Goes beyond star counts — samples actual commit messages to assess engineering discipline and team habits.

Samples 50–100 commits across top repos
Scores message quality, length, and clarity
Detects conventional commit adoption (feat/fix/chore)
Flags patterns worth asking about: 'WIP', 'fix', 'asdf'
Separate commit quality score dimension
Contextual HR insight generated from commit patterns
Feature 04

Batch Screening

Screen an entire candidate pipeline at once. Rank, compare, and export — without touching a single repo manually.

Analyze up to 20 candidates simultaneously
Side-by-side score comparison table
Automatic ranking by overall score
One-click CSV export of full batch results
Per-candidate decision badges
Priority queue — see your top candidates first

And much more

Every edge case, handled.

One-Click Reports

Shareable PDF & link exports for hiring manager briefings. No engineering required.

Interview Guide

Auto-generated interview plan with rubrics, pipeline stage, and AI-enhanced questions.

ATS Export

Push candidate data directly to Greenhouse or Lever via webhook integration.

Pull Request Analysis

View the candidate's recent PRs across public repos — real collaboration signal.

Experience Inference

Automatically determines Junior / Mid-level / Senior / Staff level from public signals.

Role Suggestion

Recommends the best-fit engineering role based on language stack and project patterns.

Privacy-First Design

Self-hosted AI, public data only. History stored in your encrypted account — deletable anytime. No third-party LLM access.

Confidence Scoring

Every decision includes a data confidence %. Low public data triggers a 'needs review' flag.

Nuanced Hire Decisions

Strong Hire / Hire / Needs Review / Insufficient Data — never a blunt reject when data is sparse.

Under 60 Seconds

Full GitHub fetch, scoring, and AI narrative delivered before your coffee cools.

Gold Standard Benchmark

Compare any candidate against your own internal engineers to calibrate the bar.

AI Transparency Layer

When AI overrides the rule engine, a banner shows exactly what changed and why.

See it in action.

Paste any public GitHub username and get a full scored report in under 60 seconds — free, no signup.

Try it Free →View PricingScoring Methodology