Clean Architecture | Daniel Arbelaez Alvarez

FocusCoach - AI Coaching with DISC Methodology

Thu, 02 Apr 2026 00:00:00 +0000

FocusCoach is a personalized coaching platform powered by artificial intelligence and the DISC personality profiling methodology, designed to help entrepreneurs (especially high-I profiles) prevent project abandonment.

Key Features

Adaptive Coaching: Personalized sessions based on the user’s DISC profile
FOCUS-SPRINT Framework: Proprietary methodology for sprint-based project management
Parking Lot: Idea management system to avoid distractions
DISC Gamification: Reward system adapted to the user’s personality profile
Active Learning: The AI learns from interactions to improve recommendations
Document OCR: Document processing with Mistral AI OCR

Technologies Used

Backend: Python 3.12, FastAPI 0.115, Pydantic v2
Frontend: Streamlit 1.41
AI: OpenAI Assistants API, LangChain
Database: SQLite + FTS5 (WAL mode)
Architecture: Clean Architecture (5 layers), DI, Railway-Oriented Programming (returns)
Quality: pytest + hypothesis, ruff, mypy –strict
Deployment: Docker (2-stage build)

Architecture

Clean Architecture implementation with 5 layers and advanced patterns:

Strategy Pattern for adapting coaching based on DISC profile
Dependency Injection for service decoupling
Railway-Oriented Programming (Result[T, E]) for error handling
28 experimental notebooks validating 8 MVPs before implementation

Achievements

46 tests passing with strict type checking (mypy –strict)
8 validated MVPs in notebooks before migration to production
5 architectural layers implemented with complete DI

Impact

FocusCoach addresses a common problem among entrepreneurs: project abandonment due to lack of focus and structure. By personalizing coaching according to DISC personality, the platform offers strategies that resonate with each user’s cognitive style.

This project combines knowledge from organizational psychology (DISC methodology), software engineering (Clean Architecture), and artificial intelligence (LLMs + RAG) to create an innovative personal productivity tool.

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.