Christian Joshua
Alberto
Automation & ML-focused developer building end-to-end systems — from n8n workflows and FastAPI backends to deployed deep-learning models. Currently automating HR operations at Crowdsource Innovative Solutions Inc.
- Designed and deployed end-to-end HR automation workflows with n8n, Google Workspace APIs, PostgreSQL and Supabase — automating a 3-tier approval system, document generation, and timesheet updates for 16 employees, cutting manual processing time by ~80%.
- Built a production-grade HR web portal (HTML/CSS/JS) with real-time dashboards, live filtering, and webhook-driven integrations, eliminating repetitive manual data entry across monthly approval cycles.
- Architected and maintained an HR database & payroll system on Supabase, centralizing employee records, payroll computation, and data access.
- Built an AI-powered Resume Fit Analysis workflow (n8n + OpenAI API) that screens candidates against job descriptions and generates fit scores and summaries.
Spatial-temporal deep-learning model (TensorFlow + Keras) forecasting dengue outbreaks across NCR. Full preprocessing & optimization pipeline, deployed on GCP, Render and Vercel with API integration for real-time risk monitoring.
n8n + OpenAI workflow that screens candidate resumes against job descriptions, generating structured fit scores and summaries to accelerate HR shortlisting — built in production at CISI.
Siamese neural network in TensorFlow for signature verification and forgery detection. Performance evaluated with Euclidean-distance metrics and confusion-matrix analysis.
IoT-enabled waste-segregation system using Arduino, ultrasonic sensors and OpenCV for real-time trash classification, with ML-based material detection and GSM alerts.
End-to-end 3-tier TRF approval system on n8n, Google Workspace APIs and PostgreSQL, fronted by a live dashboard with real-time filtering that eliminates manual timesheet entry.
Support Vector Machine predicting heart-failure risk from clinical data, with cleaning, feature scaling and hyperparameter tuning for improved medical classification accuracy.
- Led internal operations across 3 committees, improving coordination and communication efficiency.
- Planned and ran workshops and training sessions, lifting member engagement by 20%.
- Streamlined processes to cut event-prep time by 25%, with 100% SDAO documentation compliance.
- Supported technical execution and promotion of 5 Google events.
- Managed Google Workspace tools (Drive, Forms), improving resource accessibility by 30%.
- Set up and maintained technical platforms for online and on-site events.