Secure systems connected through reliable infrastructure.
Linux, PostgreSQL, containers, automation, and deployment workflows.
Credible early-career experience from local, lab, test, and supervised environments.
Featured projects
Case studies shaped around deployment decisions, data flow, and maintainability.
CloudflareCaddyNext.jsFastAPIPostgreSQL
Featured deployment project
FileDrop
A temporary cross-device file-transfer application for uploading one or more files, generating a short-lived code, and downloading the files on another device.
Role
Built and deployed the application stack, backend data model, container network, reverse proxy, persistence, and deployment documentation.
Outcome
A secure, documented, multi-service deployment workflow for short-lived file transfers with private services and public HTTPS ingress.
Deployed a multi-service Docker Compose stack with PostgreSQL, FastAPI, and Next.js services kept private.
Exposed only the Caddy reverse proxy while using private Docker networking for internal services.
Configured persistent PostgreSQL and uploads volumes with health and readiness endpoints.
Created and applied Alembic migrations for PostgreSQL metadata storage.
Implemented multiple-file uploads, single-file direct download, ZIP download for multiple files, expiring transfers, configurable download limits, and automatic physical-file cleanup.
Used SSH tunnelling during development and documented deployment, database, backup, restore, and architecture procedures.
Key challenges
Separating public ingress from private service networking.
Keeping metadata, uploads, cleanup, and download limits consistent across expiring transfers.
Documenting repeatable operations without exposing secrets or infrastructure details.
Local analysis pipeline
Forensic Video Analysis and Dataset Pipeline
A modular local pipeline that extracts video metadata, representative frames, temporal information, audio information, and structured forensic features for dataset preparation.
Role
Developed pipeline modules for auditable feature extraction, validation, structured outputs, error handling, and tests.
Outcome
A repeatable dataset-preparation pipeline that supports structured forensic analysis workflows without claiming to determine authenticity on its own.
PythonOpenCVFFmpegffprobeNumPyJSONJSONLCSVpytest
Technical details
Engineering decisions
Used versioned schemas and structured logging so generated outputs can be reviewed later.
Separated metadata, frame, temporal, audio, and feature extraction steps into maintainable modules.
Added validation, error handling, and automated tests around local processing workflows.
Key challenges
Producing consistent outputs from varied source media.
Keeping forensic preparation auditable without claiming authenticity decisions.
Lab and test-environment project
Infrastructure Automation and CI/CD Lab
A test-environment workflow for repeatable build, test, and deployment stages using containerized services and GitLab CI/CD.
Role
Configured YAML pipeline stages and delivery patterns for local or test-server deployment practice.
Outcome
A practical CI/CD lab for understanding delivery stages, configuration, and deployment flow.
Defined build, test, and deployment stages in pipeline configuration.
Used containers to make local and test-server delivery more repeatable.
Kept the project clearly scoped as a lab rather than production ownership.
Key challenges
Making pipeline steps repeatable across local and test environments.
Hackfusion 2026 team prototype
VisionX
A four-member digital-evidence analysis prototype built during Hackfusion 2026.
Role
Contributed to the homepage and analysis-results interface, supported limited video-pipeline work, and helped maintain a stable working version during development-environment issues.
Outcome
The project received the Best Social Impact Award at Hackfusion 2026.