Skip to content

Court Case Management System (Judicial Platform)

Company: Beijing Jingyu Network Technology (Judicial Systems)
Role: Senior Ruby on Rails Developer
Period: Nov 2017 – Oct 2018
Domain: Judicial Informatization / Court Case Processing Systems

1. Project Summary

The Court Case Management System is a judicial case-handling platform designed to digitize, automate, and optimize the entire lifecycle of court proceedings.

The system significantly reduced manual workload for court staff by:

  • Automating document ingestion and structuring
  • Digitizing case filing and trial preparation workflows
  • Assisting judges in legal research, judgment drafting, and case review
  • Enabling knowledge reuse through structured archival
Project Flow: Intelligent Judicial Assistance System

2.1 Case Filing

Traditional Workflow

  • Plaintiffs submit complaints, identity documents, and evidence (PDFs, scans, images).
  • Filing clerks manually enter case details (category, parties, amounts).

Pain Points

  • Inconsistent formats and high-volume repetitive data entry.

System Solution

  • OCR & Structured Parsing: Extract text and identify key entities (Plaintiff, Monetary amounts, etc.).
  • Intelligent Recommendation: Based on historical data.

2.2 Pre-trial Review

Traditional Workflow

  • Judges manually search laws and similar cases based on personal experience.

Pain Points

  • Time-consuming and inconsistent document review.

System Solution

  • Automatic Similar Case Retrieval: Vectorize key facts to find the Top 5 similar historical cases.

2.3 Trial & Hearing

Traditional Workflow

  • Manual note-taking and high effort in organizing Trial records.

System Solution

  • Template-Based Hearings: Predefined question templates by case type.

2.4 Judgment Drafting

Traditional Workflow

  • Even simple cases take 1–2 hours for drafting and legal research.

System Solution

  • Automated Draft Generation: Combines current facts with patterns from similar judgments.
  • Key Design Principle: System provides reference versions only; no auto-generated final verdicts.

2.5 Judge Review & Confirmation (Human-in-the-loop)

  • Full Control: All system-generated content is fully editable and traceable.
  • Auditability: Records machine-generated vs. judge-modified content for legal accountability.

2.6 Delivery & Knowledge Feedback

  • Archiving: Final judgments are automatically stored in structured form.

User Flow Diagram

3. System Architecture

Core Technology Stack

  • Backend: Ruby on Rails (Full-stack)
  • Web Server: Nginx (Reverse Proxy)
  • Relational Database: MySQL (Case metadata, parties, workflow states)
  • Document Storage: MongoDB (Structured legal documents)
  • Search & Similarity Engine: Apache Solr
  • Cache & Queue: Redis
  • Async Processing: Sidekiq
  • Deployment: On-Site court environment

Architecture Characteristics

  • Nginx routes requests to Rails backend
  • Sidekiq handles heavy background jobs:
  • OCR processing
  • Document structuring
  • Similar case indexing
  • MongoDB stores structured document data
  • Solr supports full-text search and similarity matching
  • Redis used for caching and job queues

User Flow Diagram

4. Technical Challenges & Solutions

Large-Scale Document Processing

Challenge

  • Courts process massive volumes of unstructured legal documents daily.

Solution

  • OCR + background processing via Sidekiq
  • Document parsing decoupled from request lifecycle

Challenge

  • Efficiently finding legally relevant similar cases.

Solution

  • Hybrid approach:
  • Text vectorization
  • Rule-based constraints (case type, amount)
  • Solr optimized for legal text search and ranking

Judicial Reliability & Trust

Challenge

  • Automation must not override judicial authority.

Solution

  • Human-in-the-loop workflow
  • Full editability and source traceability
  • No automated verdict generation

5. My Responsibilities & Achievements

Responsibilities

  • Led on-site requirement analysis with court staff
  • Translated judicial workflows into technical designs
  • Oversaw system deployment within court environments
  • Managed requirement changes and iteration cycles
  • Coordinated developers, stakeholders, and court officials

Key Achievements

  • Led end-to-end project delivery across multiple court deployments, ensuring on-time launch and stable operation in on-premise judicial environments
  • Drove requirement analysis and process modeling by working closely with judges and court staff, translating legal workflows into actionable system designs
  • Coordinated cross-functional teams (backend, frontend, QA, deployment) to deliver iterative releases aligned with contract and regulatory requirements
  • Established effective communication and delivery mechanisms between legal stakeholders and engineering teams, reducing requirement ambiguity and rework
  • Built and managed the development team, overseeing task breakdown, progress tracking, and delivery quality throughout the project lifecycle