Small web features · Python fixes · AI-assisted development
Hanoi, Vietnam · English / Vietnamese
I build small, inspectable app features. A useful starting task is one form with validation, a clear preview and an export, delivered with source files, tests and run instructions. Send a field list, mockup or reproducible bug; we can agree a bounded scope and fixed price before work starts.
Try the working samples · Discuss one component
| Project | What to inspect |
|---|---|
| Interactive frontend samples | Try form validation and retained input, source-preserving review, and text/JSON exports. Nine independent HTML/CSS/JavaScript prototypes with synthetic examples and documented checks. |
| Merged Python contribution | Fixed a bug that broke style lookups containing quotation marks in python-docx-ng. Includes regression tests; reviewed and merged by the maintainer. |
| AI Systems Lab | Independent AI-assisted local PDF copilot with source-page citations, tenant controls and durable human approvals. Reproducible checks, historical local-model evidence and documented limits; production acceptance remains incomplete. |
| Webhook-to-CRM reliability sample | Python and SQLite signature verification, duplicate/conflict handling, bounded retries and operator recovery. Uses synthetic events and a simulated CRM. |
| Vehicle tracking workflow | Telegram input, license-plate recognition integration, notifications and spreadsheet tracking. Offline boundary checks are available separately. |
My background combines Human Resource Management coursework at RMIT University Vietnam with import/export documentation work. That operational experience shapes how I approach software: understand the process, define its failure cases and build a useful system.
I use AI-assisted development and publish reproduction instructions, test evidence and limitations with my projects. Current work spans Python, JavaScript/TypeScript, SQL, FastAPI, webhooks and document workflows.
Open to automation, integration and implementation roles.
CV · CV PDF · LinkedIn · Email
Automotive service-event case study · Additional project evidence
These are independent samples, personal projects and open-source work. Each repository states its tested scope; paid-client outcomes and production readiness are not inferred from passing local checks.