Hugging Face's logo

deeep-sec
/
campus-placement-engine

Model card Files
xet
 Community
campus-placement-engine / docs /OFFICER_USER_GUIDE.md
deeep-sec's picture
deeep-sec
feat: institutional campus placement eligibility & interview scheduling engine
788eab4 verified
preview code
|
Raw History Blame Contribute Delete
7.11 kB

Institutional Placement Engine: Placement Officer & Coordinator Guide (v1.0.0)

Welcome to the Campus Placement Engine

The Institutional Placement Engine is a purpose-built, 100% offline platform designed to orchestrate complex campus recruitment seasons. It operates directly on institutional laptops or local servers without requiring an internet connection, cloud subscriptions, or Docker.

1. Quick Launch & Single-Port LAN Deployment

  1. Launch the Engine:
    • Double-click start.bat (Windows) or execute ./start.sh (Linux/macOS) in the root directory.
    • Alternatively, execute:
      python scripts/lan_launcher.py
  2. Access the Application:
    • Local Officer Console: http://localhost:8000
    • Candidate & Recruiter LAN Access: The launcher automatically prints your active local IP (e.g. http://192.168.1.45:8000). Distribute this URL to students across the college Wi-Fi.
  3. Default Credentials:
    • Placement Officer: Username: officer | Password: officer123
    • Placement Coordinator: Username: coordinator | Password: coord123
    • Sample Student: Roll No: PIPE_001 | OTP: 123456

2. Ingesting Candidate Rosters (CSV / XLSX)

  1. Open the Students & Importer tab (/students).
  2. Drag and drop your college master spreadsheet (.xlsx or .csv).
  3. Automatic Column Matching: The engine auto-detects common header variants (Roll Number, RegNo, CGPA, 10th %, Backlogs). Manually map any unmatched columns.
  4. Dry-Run Audit: Click Audit & Preview. The engine checks:
    • Scale conversions (4.0 to 10.0 CGPA).
    • Branch aliases (e.g., "CompSci" -> "Computer Science and Engineering").
    • Invalid dates and formatting issues.
  5. Review the row-level audit table. If satisfactory, click Commit Ingestion to save records.

3. Extracting Drive Requirements from Circulars

  1. Open the Extract Drive tab (/create_drive).
  2. Provide the announcement document:
    • Paste raw email text directly, OR
    • Upload a PDF circular (digital text and tables are automatically parsed), OR
    • Upload a Word bulletin (.docx).
  3. Click Extract Requirements.
  4. Side-by-Side Verification:
    • Left side shows the raw document with highlighted source sentences.
    • Right side displays parsed fields (CGPA cut-off, branch list, backlog policy, package/CTC, interview rounds).
    • Confidence badges indicate extraction certainty. Confirm any low-confidence fields and click Create Drive & Ruleset.

4. Configuring Rules & Policy Toggles

  1. Rule Editor (/rule_editor):
    • Visual forms allow adjusting cut-offs, branch aliases, maximum allowed backlogs, and gap-year ceilings without code.
    • Live candidate counter shows the number of eligible candidates updating in real time.
  2. Offer Policy Settings (/policies):
    • Select an institutional policy model (Strict One-Offer, Progressive 1.5x Multiplier, or Tiered Quota).
    • Configure institutional interpretation toggles:
      • Is Dream threshold inclusive? (e.g. is 10.0 LPA exactly treated as Dream?).
      • Do off-campus offers count against quota?
      • Mandatory cooling window duration (days).

5. Running Batch Eligibility & Audit Verification

  1. Open the Eligibility Runs tab (/eligibility).
  2. Select the recruitment drive and click Run Batch Eligibility.
  3. The engine evaluates all candidates in milliseconds:
    • Categorizes students into Eligible, Ineligible, and Policy Blocked.
    • Generates single-line plain-English explanations for candidates.
    • Calculates a cryptographic SHA-256 snapshot hash ensuring audit reproducibility.
  4. Export the resulting shortlist or proceed directly to scheduling.

6. Generating Clash-Free Schedules (OR-Tools CP-SAT)

  1. Open the Schedule Grid tab (/schedule).
  2. Select participating drives, available rooms (auditoriums, labs, interview cabins), and date windows.
  3. Click Generate Clash-Free Schedule.
  4. The CP-SAT solver computes an optimal timetable in seconds:
    • Zero Room Overlaps: No two companies book the same hall simultaneously.
    • Zero Panel Clashes: Interviewers are assigned non-overlapping student slots.
    • Zero Candidate Clashes: Students appearing for multiple drives are guaranteed collision-free slots with walking buffers between campus blocks.
  5. The UI renders an interactive Rooms $\times$ Time grid. Drag and drop slots if manual adjustments are needed; the engine instantly flags any new clashes.

7. What-If Simulation & Publish Workflow

  1. Open the What-If & Publish tab (/what_if).
  2. Before publishing changes, simulate prospective adjustments:
    • E.g., lowering CGPA cut-off from 7.0 to 6.5, adding an interview panel, or marking a room unavailable.
  3. The engine computes an instant Impact Report:
    • Generates a data-driven headline: "Lowering cut-off adds 38 students; 6 clash with Northwind test."
    • Previews schedule movements required to restore zero-clash integrity.
  4. Click Publish Official Schedule to lock in an immutable version snapshot.
  5. In case of unexpected changes, click Rollback to Previous Version for instant state recovery.

8. Result Entry & Automated Enforcement ChangeSets

  1. Open the Result Entry tab (/results).
  2. Enter company hiring outcomes (offers accepted, rejected, or revoked).
  3. The engine computes a live Enforcement ChangeSet:
    • Identifies students who now exceed offer quotas.
    • Automatically cancels downstream test/interview reservations for blocked students.
    • Frees up interview slots for waiting-list candidates.

9. Resolving Candidate Disputes

  1. Open the Dispute Explainer tab (/dispute).
  2. When a student questions why a peer was permitted to interview while they were blocked:
    • Select the recruitment drive.
    • Select Candidate A (allowed) and Candidate B (blocked).
  3. The engine generates an objective side-by-side justification citing the exact rule AST nodes and policy clauses, completely eliminating subjective allegations.

10. Data Protection, Encrypted Backups & Institutional Wipe

  1. Air-Gap Verification:
    • Visit the Offline Audit tab (/offline_audit) to verify that 0 external network requests are being made.
  2. Encrypted Backups:
    • Export an AES-256 encrypted database backup via the CLI:
      python scripts/backup_restore.py backup --passphrase "InstitutionalSecretKey"
  3. Restore Database:
    • Restore state on a clean machine:
      python scripts/backup_restore.py restore my_backup.enc --passphrase "InstitutionalSecretKey"
  4. Institutional Wipe:
    • At the conclusion of the academic year, wipe all student records and credentials by issuing the cryptographic wipe confirmation: "CONFIRM_INSTITUTIONAL_WIPE".