# Enterprise High-Volume Structured Hiring & ATS Lab

An audit-ready portfolio simulation of a governed hiring system for Apex Precision Dynamics Ltd. (APD), a fictional 70-person precision-manufacturing firm in Aligarh, India. The enterprise edition demonstrates how structured selection, automation, privacy, fairness monitoring, and recruiter operations can remain traceable across **4,000 synthetic candidates**.

## Executive summary

Five requisition families generated 4,000 applications, 986 automated-knockout progressions, 500 fully assessed finalists, and 120 hires. Selection uses a governed 40/40/20 composite of work sample, structured BARS interview, and job-knowledge evidence. The reconciled operating view is **3.0% conversion**, **28.5 average days to fill**, **91.8% feedback-SLA adherence**, and a **0.87 adverse-impact ratio** at the knockout-progression gate.

Candidate CAND-2026-0013 is the halo-effect control. A subjective impression of 4.60 did not override a governed composite of 3.92; the +0.68 gap triggered evidence review and the candidate was not hired.

All people, identifiers, dates, scores, and events are fictional. Demographic cohorts exist only for aggregated fairness monitoring and are inaccessible to decision-makers.

## Explore the lab

- [Open the enterprise project experience](index.html)
- [Open the high-volume analytics dashboard](dashboard.html)
- [Open the five-slide case presentation](slides.html)
- [Download the five-page PDF case study](Structured_Hiring_and_ATS_Architecture_Case_Study.pdf)
- [Open the responsive phone case-study source](mobile-case-study.html)
- [Read the five-page phone-friendly portrait edition](Structured_Hiring_and_ATS_Architecture_Mobile_Case_Study.pdf)

Both dashboard pages embed all three source datasets as JSON. Search, filtering, 25/50-row pagination, charts, and scorecards run directly from the local file. Custom CSS preserves the application shell if the visual CDN libraries are unavailable.

## Governed KPIs

| KPI | Reconciled value | Source of truth |
|---|---:|---|
| Candidates evaluated | 4,000 | Candidate row count |
| Applied-to-hired conversion | 3.0% | 120 hires / 4,000 applicants |
| Average time to fill | 28.5 days | Mean of 34.0, 22.0, 30.0, 26.0, and 30.5 |
| Interviewer SLA adherence | 91.8% | 1,836 of 2,000 evaluations at or below 48 hours |
| Knockout-progression AIR | 0.87 | (362/1,600) / (624/2,400) = 0.870 |

The 4/5ths result is a monitoring signal rather than proof of fairness or lawful practice. The larger cohort improves precision but does not repair invalid criteria, biased measurement, job-family confounding, or poor data quality.

## Requisition families

| Requisition | Applicants | Shortlisted | Assessed | Hired |
|---|---:|---:|---:|---:|
| Senior Precision Engineer · G4 | 800 | 197 | 100 | 10 |
| CNC Precision Machinist Trainee · G1 | 1,600 | 395 | 200 | 60 |
| Quality Assurance Specialist · G3 | 600 | 148 | 75 | 15 |
| People Operations Specialist · G2 | 400 | 98 | 50 | 10 |
| Supply Chain & Logistics Associate · G2 | 600 | 148 | 75 | 25 |
| **Total** | **4,000** | **986** | **500** | **120** |

## Repository map

### Data

- [Synthetic requisitions](synthetic_requisitions.csv)
- [Synthetic candidates](synthetic_candidates.csv)
- [Synthetic interviews](synthetic_interviews.csv)

### Evidence, design, and governance

- [Research foundation](RESEARCH_FOUNDATION.md)
- [Requisitions and roles](REQUISITIONS_AND_ROLES.md)
- [Structured interview rubrics](STRUCTURED_INTERVIEW_RUBRICS.md)
- [ATS workflow and RACI](ATS_WORKFLOW_AND_RACI.md)
- [Selection validity model](SELECTION_VALIDITY_MODEL.md)
- [Compliance and fairness matrix](COMPLIANCE_AND_FAIRNESS_MATRIX.md)
- [RBAC and privacy matrix](RBAC_AND_PRIVACY_MATRIX.md)
- [UAT test register](UAT_TEST_REGISTER.md)
- [Methodology and limitations](METHODOLOGY_AND_LIMITATIONS.md)
- [Data dictionary](DATA_DICTIONARY.md)

### Career assets

- [CV project entry](CV_PROJECT_ENTRY.md)
- [LinkedIn post](LINKEDIN_POST.md)
- [Interview guide](INTERVIEW_GUIDE.md)

## Reproduction

1. Clone or download the portfolio repository.
2. Open `index.html`, `dashboard.html`, or `slides.html` in a modern browser.
3. Inspect the three CSV files for row-level evidence.
4. Run `python3 98_Maintenance/tests/test_structured_hiring_ats_lab.py` from the repository root.

The acceptance suite checks the exact inventory, schemas, row counts, references, scoring arithmetic, KPI reconciliation, privacy patterns, enterprise pagination controls, embedded JSON parity, and five-slide contract.

## Interpretation boundary

This is a systems-design and analytics work sample, not a validated production selection instrument, legal opinion, or employment recommendation. Production use requires local job analysis, candidate accessibility review, criterion validation, security and load testing, employment-law review, recruiter-capacity planning, and ongoing subgroup monitoring.