LegalTech | Daniel Arbelaez Alvarez

DisciplinaJudicial AI - Judicial Discipline Process Guidance

Fri, 06 Mar 2026 00:00:00 +0000

DisciplinaJudicial AI is an artificial intelligence assistant specialized in disciplinary processes for Colombian judicial servers, providing guidance on the disciplinary regime, offenses, sanctions, and rights.

Key Features

Disciplinary Regime: Guidance on the regulatory framework for judicial server discipline
Offense Classification: Assistance on the categorization and severity of disciplinary offenses
Disciplinary Procedure: Support on stages and deadlines of the process
Sanctions: Information on types of sanctions and their implications
Rights of the Investigated: Guidance on procedural guarantees

Technologies Used

Backend: Python, OpenAI API
Frontend: Streamlit
AI: Language model with a knowledge base in judicial disciplinary law
Deployment: Streamlit Cloud

Impact

DisciplinaJudicial AI facilitates access to information about the judicial disciplinary regime, helping judicial servers understand their obligations, rights, and the consequences of failing to fulfill their functional duties.

This assistant was born from the need to provide accessible guidance in a highly specialized area of law, where information is often fragmented and difficult to interpret for officials without deep legal training.

SIRAL - Workplace Harassment Registration System

Thu, 15 Jan 2026 00:00:00 +0000

SIRAL is a comprehensive platform for managing workplace harassment complaints in the Colombian Judicial Branch, developed in compliance with Law 1010/2006, Law 1581/2012 (Habeas Data), Resolution 3461/2025, and the General Archives Law.

Key Features

Complaint Filing: 4-step registration workflow with mandatory identification of involved parties
Role-Based Access Control (RBAC): 5 differentiated roles (Superadmin, CCL Admin, CCL Member, Official, Public)
Resolution Workflow: State management with 6-month deadline per Resolution 3461/2025
Chain of Custody: Immutable SHA-256 hash for digital evidence
Digital Notifications: Automated communication to the alleged harasser (Law 2213/2022)
Committee Minutes: Recording and management of Workplace Coexistence Committee minutes
Anonymized Statistics: Public reports without personal data
Document Retention: 20-year preservation in accordance with the General Archives Law

Technologies Used

Backend: Python 3.12+, FastAPI, Pydantic v2, SQLAlchemy 2.0 (async)
Database: PostgreSQL 16
Frontend: Streamlit (10 pages + sidebar + timeline + charts)
Security: JWT (python-jose, 15 min expiration), bcrypt, RBAC, OWASP Top 10
Infrastructure: Docker Swarm, EasyPanel, Traefik (HTTPS), GHCR
CI/CD: GitHub Actions (6 jobs: lint + test + build + deploy)
Quality: pytest, ruff, mypy, bandit

Architecture

Layered architecture following Clean Architecture:

Core: Domain entities, value objects, enums
Application: Use cases, DTOs (frozen dataclasses), service ports
Infrastructure: Persistence, security, notifications, storage
Interfaces: FastAPI routes (9 endpoints + Swagger), Streamlit GUI

Achievements

283 tests passing at 100% with complete quality gates (pytest + ruff + mypy + bandit)
99 Python source files in src/
6 use cases implemented (file complaint, change status, query, list, detail, minutes)
55 audit findings resolved at 100% (5 review agents)
10 planning documents (P00-P09) and 4 sprints completed
Security: OWASP Top 10 remediated, CSP, HSTS, path sanitization

Impact

This system fully digitizes the workplace harassment management process in the Judicial Branch, ensuring regulatory compliance, traceability of every action, and protection of the personal data of involved parties.

SIRAL arose from the need to formalize and systematize the workplace harassment complaint process at the Judicial Services Center of the Antioquia Sectional, replacing manual processes prone to errors and delays with a robust, auditable technology solution.

Sherlock-docs - Intelligent Legal Document Processing

Mon, 01 Dec 2025 00:00:00 +0000

Sherlock-docs is an intelligent processing system for judicial documents (tutelas and habeas corpus) that combines OCR, named entity recognition (NER), and duplicate detection. All processing is 100% local, with no data sent to external APIs.

Key Features

Hybrid OCR: PaddleOCR + Tesseract for scanned and digital documents
Legal NER: Entity extraction with SpaCy, F1 score of 85.3% validated by humans
Multi-Level Duplicate Detection: SHA-256 hash, LZJD fuzzy hash, TF-IDF, Sentence-Transformers
Advanced Search: SQLite FTS5 full-text search with advanced filtering
Complete REST API: 22 FastAPI endpoints with JWT authentication and Swagger documentation
Graphical Interface: 9 Streamlit pages for interactive document management
CLI: 20 commands for batch operations
Active Learning: Interface for human validation of NER entities
Export: Excel report generation

Technologies Used

Backend: Python 3.12.4, FastAPI, Pydantic
OCR: PaddleOCR + Tesseract
NLP/NER: SpaCy (Spanish legal model)
Database: SQLite + FTS5 (33 columns, 11 indexes, WAL mode)
Frontend: Streamlit (9 pages)
Architecture: Layered Clean Architecture, Result-Oriented Programming (returns)
Security: JWT, RBAC, rate limiting, security headers
Deployment: Docker, Easypanel, automated GitHub Webhook

Architecture

5 layers with ServiceContainer (14 lazy-loaded properties):

Core: Entities, value objects
Application: Use cases, DTOs, ServiceContainer
Infrastructure: OCR, NER, deduplication, logging
Persistence: SQLite + FTS5 with ISP ports
Interfaces: Streamlit GUI, CLI (20 commands), FastAPI REST (22 endpoints)

Achievements

1,832 tests (1,770 unit + 62 integration) with 87% coverage
NER F1 85.3% with data validated by human operators
22 REST endpoints documented and functional
16/16 security SECs + 4 SEC-API implemented
9 sprints completed (S23-S28) with 30 planning documents
mypy 0 errors in strict mode across 130 files
Certified SDD audit: 39 conforming, 0 defects
Code quality: 9.1/10

Impact

Sherlock-docs processes ~100 documents/day with a performance target of <2 minutes for 20-page documents and <100ms for digital documents, eliminating manual classification of judicial documents and automatically detecting duplicate filings.

This system was born from the need to automate the registration and classification of tutelas and habeas corpus in the Colombian judicial system, where duplicate detection and accurate extraction of procedural party information are critical to judicial office efficiency.

Docassemble Expert - Legal Document Automation

Thu, 25 Sep 2025 00:00:00 +0000

Docassemble Expert is a specialized assistant for the Docassemble framework, designed to help developers, lawyers, and organizations create automated web interviews and legal document generation systems.

Key Features

YAML Expertise: Advanced knowledge of Docassemble’s YAML syntax
Interview Design: Patterns for creating guided web interviews
Document Generation: Automation of legal document creation
Best Practices: Guidance on design patterns and form architecture
Conditional Logic: Assistance with complex interview flows and branching

Technologies Used

Framework: Docassemble (open-source expert system)
Configuration Language: YAML
Backend: Python
Documentation: Markdown

Impact

Docassemble Expert facilitates the adoption of document automation tools in the legal sector, reducing the technical barrier for lawyers and organizations seeking to digitize their document generation processes.

This project supports the vision of democratizing legal automation, enabling legal professionals to create their own document generation tools without requiring deep programming knowledge.

TYBABot - Virtual Assistant for Justicia XXI Web

Fri, 19 Sep 2025 00:00:00 +0000

TYBABot is a specialized virtual assistant for Colombia’s Justicia XXI Web (Tyba) system, developed with Streamlit and the OpenAI API. It provides guidance on electronic judicial management to officials and system operators.

Key Features

Document Management: Guidance on e-judicial documents and their handling in Tyba
Judicial Proceedings: Guide on registering and querying procedural actions
Petitions: Assistance in managing electronic petitions (memoriales)
Notifications: Guidance on the digital notification system
Electronic Signatures: Support for judicial electronic signature processes

Technologies Used

Backend: Python, OpenAI Assistants API v2
Frontend: Streamlit
AI: Language model with specialized knowledge base on Tyba
Deployment: Streamlit Cloud

Impact

TYBABot reduces the learning curve for judicial officials interacting with the Tyba system, providing contextualized responses based on the official system documentation and best practices for electronic judicial management.

This assistant was born from direct experience working with the Justicia XXI Web system in the Judicial Branch, identifying the most frequent difficulties officials face when adopting electronic document management tools.

TutelaBot - Fundamental Rights Assistant

Thu, 18 Sep 2025 00:00:00 +0000

TutelaBot is a virtual assistant specialized in Colombian constitutional mechanisms for the protection of fundamental rights: the Accion de Tutela (constitutional protection action) and the Derecho de Peticion (right to petition).

Knowledge Base

Political Constitution of Colombia 1991: Relevant articles on fundamental rights
Decree 2591 of 1991: Regulation of the Accion de Tutela

Key Features

Tutela Guidance: Assistance on requirements, procedures, deadlines, and scope of the Accion de Tutela
Right to Petition: Support on how to exercise the right to petition before public and private entities
Legal Foundation: Responses grounded in current constitutional regulations
Accessible Language: Clear explanations for citizens without legal training
Conversational Interaction: Natural dialogue to resolve legal questions

Technologies Used

Backend: Python, OpenAI API
Frontend: Streamlit
AI: Language model with Colombian constitutional knowledge
Deployment: Streamlit Cloud

Impact

TutelaBot democratizes access to information about fundamental rights protection mechanisms, enabling any citizen to understand how to exercise their constitutional rights without the need for prior legal counsel.

This project stems from the conviction that technology can reduce the justice access gap by providing basic legal guidance on the most widely used mechanisms by Colombian citizens to protect their fundamental rights.

API Consumer - Judicial Data Extraction

Sun, 01 Jun 2025 00:00:00 +0000

Rama Judicial API Consumer is an ETL (Extract, Transform, Load) system that consumes the Colombian Judicial Branch’s public REST API to extract judicial process information, including PDF documents, and transform it into structured text format for analysis.

Key Features

REST API Consumption: Integration with the Judicial Branch’s public API
PDF Extraction: Automatic download and parsing of procedural documents
Rate Limiting: Request throttling to respect API limits
Text Transformation: Conversion of PDFs to structured plain text
Error Handling: Automatic retries and failure logging
Caching: Local storage of responses to avoid redundant requests

Technologies Used

Language: TypeScript
Runtime: Node.js
HTTP Client: axios
Rate Limiting: p-limit
PDF Parsing: pdf-parse
Output: Structured TXT files

Impact

This system enables automated extraction of judicial data for statistical analysis and data mining, facilitating research on procedural patterns and report generation based on public information from the Judicial Branch.

Developed as a data pipeline to feed statistical analyses and intelligence tools on Colombian judicial processes, demonstrating integration capabilities with government APIs and document processing.

DocumentGenerator - Bulk Document Generation

Sun, 01 Jun 2025 00:00:00 +0000

DocumentGenerator is a bulk document generation system that uses Word templates and Excel spreadsheet data to produce large volumes of standardized documents with concurrent processing.

Key Features

Bulk Generation: Creation of hundreds of documents from a single template and XLSX data
Concurrent Processing: ThreadPoolExecutor for parallel generation
Flexible Templates: python-docx-template engine with Jinja2 syntax
YAML Configuration: Complete parameterization without code modification
Structured Logging: JSON logging for every operation
Error Management: Robust per-document error handling

Technologies Used

Language: Python 3.9+
Templates: python-docx-template (Jinja2 for Word)
Data: pandas (XLSX reading)
Concurrency: concurrent.futures (ThreadPoolExecutor)
Configuration: YAML
Logging: Structured JSON

Impact

DocumentGenerator automates batch creation of official documents, eliminating manual production of repetitive documents and reducing transcription errors in environments that require generating large volumes of standardized communications.

This project was born from the need to generate official documents in bulk within judicial offices, where each document shares a common structure but differs in the specific data of the parties involved.

GexCom - Judicial Communications Management System

Mon, 01 Jan 2024 00:00:00 +0000

GexCom is a communications management system developed to streamline and improve the accuracy of procedural party communications within the judicial system.

Key Features

Communication Automation: Automated management of hearing citations
API Integration: Message delivery through specialized APIs
Format Management: Automated creation of referrals and postal mail formats
Intelligent Storage: Database for relevant notification information
Process Standardization: Uniformization of administrative procedures

Technologies Used

Backend: Python 3
Database: SQLite
Version Control: Git and GitHub
IDE: Visual Studio Code
APIs: Integration with messaging services

Achievements

Significant improvement in the efficiency and accuracy of communications management, specifically in the citation of parties to scheduled hearings and communications requested by Judicial Branch dependencies.

This project arose from the need to optimize the exhaustive operations in the judicial communications process, providing a technology solution that standardizes and automates the most critical administrative processes in the Colombian judicial system.

Electronic Case File Management

Mon, 01 Aug 2022 00:00:00 +0000

Electronic Case File Management is a system developed to automate the assembly of electronic case files across all dependencies of the Colombian Judicial Branch. The project has evolved through two major versions.

Version 1 - Desktop Application (2022)

MVC Architecture: Model-View-Controller pattern
Graphical Interface: tkinter for judicial operators
PDF Processing: pypdf2 for procedural document manipulation
Excel Integration: xlwings and pandas for data manipulation
Language: Python 3.9.6

Version 2 - RDA + Streamlit (2024)

The second version combines Robotic Desktop Automation (RDA) with modern technologies:

Web Frontend: Streamlit deployed in the cloud (Live App)
RDA: Robotic Desktop Automation for interaction with legacy systems
Advanced Processing: Integration with modern document processing technologies

Key Features

Full Automation: Eliminates manual case file assembly operations
PDF Processing: Automated organization, renaming, and structuring
Judicial Integration: Designed for the specific workflow of the Judicial Branch
Web Access: Version 2 accessible from any browser
Ease of Use: Intuitive interface for judicial operators without technical knowledge

Achievements

~200% reduction in time for electronic case file formalization, significantly optimizing operational efficiency in judicial dependencies nationwide.

Impact

This project assists the manual operations of electronic case file assembly, automating repetitive processes and drastically reducing processing times. Version 2 amplifies this impact by being accessible from the cloud.

The evolution of this system reflects the author’s technological growth: from a desktop application with tkinter to a modern web solution with Streamlit and RDA, always maintaining focus on solving real problems for Colombian judicial operators. Repositories: Version 1 | Version 2