Software Developer | Full Stack & Backend | Systems Integration
I develop web applications, REST APIs, backend services, and system integrations. My focus is turning concrete requirements into maintainable software: from interfaces and database changes to authentication, deployment tooling, and automated tests.
Completed project · Solution requested by UNIDA
I built a Full Stack feedback solution that connects with Canvas LMS through LTI 1.3. It uses a React interface, Node.js/Express REST services, and PostgreSQL to support academic feedback workflows, including AI-assisted generation and instructor review before delivery.
What I built and solved:
- Full Stack features: Implemented and integrated administration-panel improvements, REST endpoints, and PostgreSQL migrations for system configuration and AI-provider settings.
- LMS integration and authentication: Fixed Canvas integration issues involving OAuth2, token handling, file previews, controllers, middleware, and backend services.
- Deployment tooling: Automated installation and maintenance tasks using Docker and HTTPS/TLS, addressing filesystem permissions, non-privileged execution, and compatibility across Windows and Linux environments.
- Software quality: Added regression tests and diagnostic tools to reproduce integration issues, verify fixes, and make local setup more consistent.
- Feedback workflow: Connected the application to a process where generated academic feedback can be reviewed and approved by the instructor before submission.
Stack: React · JavaScript · Node.js · Express · REST APIs · PostgreSQL · Redis · Docker · LTI 1.3 · OAuth2 · HTTPS/TLS
Evidence: Source code · System architecture · Testing strategy
Scope: The solution is finished for its local execution environment. This is not a claim of an institution-wide production deployment.
- Web applications: Interfaces connected to APIs and business workflows.
- Backend services and integrations: REST endpoints, authentication, database-backed features, and connections between systems.
- Automation and tooling: Repeatable processing workflows, development environments, installation tooling, and diagnostics.
- Testing and maintainability: Regression coverage, modular implementation, troubleshooting, and reproducible setup.
Python application that processes authorized Zoom recordings and YouTube content, tracks interrupted transcription jobs, and exports TXT, SRT, and VTT using faster-whisper.
React and TypeScript institutional website demonstration for a fictional school, focused on responsive design, accessibility, SEO, security headers, CI, and automated checks.
- Languages: JavaScript, TypeScript, Python, SQL
- Full Stack and backend: React, Node.js, Express, FastAPI, REST APIs
- Data and infrastructure: PostgreSQL, Redis, Docker, Git/GitHub
- Engineering: Systems integration, modular architecture, automated testing, GitHub Actions
Additional repositories: Clinical Intelligence Platform · LithiumScope
