I build reliable digital products from the user interface to APIs, databases, testing, and deployment.
I'm a CS graduate with experience shipping full-stack applications, REST APIs, automated pipelines, cloud-deployed systems, and AI/computer-vision integrations.
- I work across the stack and choose technologies based on the problem
- My experience includes C#/.NET + Vue.js, Ruby on Rails + PostgreSQL, Python + Flask, Java/Android, and Swift/iOS.
- I enjoy building complete web experiences for product teams and small businesses
- I use Claude, Cursor, and other AI development tools to accelerate code exploration, prototyping, debugging, testing, and documentation
- I care about usable and accessible software
- Based in Michigan and open to relocation or remote work
- Full-stack web applications with responsive interfaces, REST APIs, authentication, databases, and third-party integrations
- Business websites and internal tools including dashboards, filtered data views, export workflows, and content-driven experiences
- Cloud and automation systems using Python, Docker, AWS, CLI tooling, and reproducible processing pipelines
- AI-enabled products using LLM workflows, computer vision, real-time inference, and cross-device communication
- Reliable software supported by unit and integration testing, code coverage, accessibility checks, peer review, and structured debugging
Built and shipped full-stack functionality for client applications using C#/.NET and Vue.js. My work included REST API development, responsive UI components, data-heavy views, filtering and export workflows, QA, peer review, and deployment support. I also used Fine Code Coverage in Visual Studio to inspect unit and integration test coverage and identify untested paths.
LSA SpaceReady — University of Michigan
Contributed to a production Rails application used to coordinate nightly classroom inspections across 300+ rooms. I built and deployed core features, restructured the PostgreSQL schema, rewrote queries for 60% faster performance, helped reduce nightly check time by 40%, and identified or resolved 15+ workflow and accessibility issues. Accessibility work included keyboard navigation, labels, semantic structure, and color contrast using WAVE.
Everyday Agent (Wearable AI Assistant) — NTT DATA Capstone (Private)
Architected a Flask REST API with 10+ endpoints for NLP queries, location prediction, and bidirectional device synchronization. Integrated SQLite, Firebase, and Google Calendar, and implemented real-time YOLOv8 object detection with OpenCV and PyTorch on a Raspberry Pi Zero 2 W. Presented the system architecture, design tradeoffs, and live functionality to NTT DATA stakeholders.
Scientific Workflow Automation — ATTMOS (Private)
Built automated Python pipelines that connected multiple computational tools into reproducible end-to-end workflows. Containerized GPU-accelerated applications with multi-stage Docker builds, deployed them to AWS EC2 with CUDA, and created CLI wrappers that standardized configuration, inputs, outputs, and evaluation across processing stages.
Spartan Planner — Campus Event Scheduler
A Java Android application that helps MSU students discover, create, and organize campus events. The app uses Firebase Authentication, Cloud Firestore for event and user data, real-time synchronization across devices, and SharedPreferences for lightweight local state such as login status.
| Project | What it demonstrates | Technologies |
|---|---|---|
| Photos Cleaner | Mobile UI development and an intuitive swipe-to-delete photo-cleanup experience | Swift, iOS |
| Sparty Sudoku | Cross-platform desktop development, modular OOP, XML parsing, unit testing, and build configuration | C++, wxWidgets, XML, CMake |
| Portfolio Website | Responsive personal-site development and interactive navigation | HTML, CSS, JavaScript, Bootstrap, FullPage.js |
| Aquarium App | Advanced object-oriented design, dynamic behaviors, and desktop GUI development | C++, wxWidgets |
I treat AI as an engineering accelerator and not a substitute for understanding the system.
- Explore: use Claude or Cursor to map unfamiliar code, dependencies, and data flow.
- Plan: break work into small, testable changes and identify risks before implementation.
- Build: prototype UI, API, data, or automation work while keeping the solution consistent with the existing architecture.
- Verify: review generated code, run unit and integration tests, inspect coverage, test edge cases, and reproduce behavior manually.
- Refine: improve readability, accessibility, performance, documentation, and maintainability before considering the work complete.
- A redesigned personal portfolio focused on full-stack case studies and measurable engineering outcomes
- Reusable, production-minded website templates for small businesses
- More public end-to-end projects that demonstrate frontend, backend, database, deployment, accessibility, and AI integration in one place
I'm interested in full-time software engineering roles, collaborative product work, and select projects for businesses that need an end-to-end web presence or custom application.
- LinkedIn: linkedin.com/in/palinaskakun
- Email: palinaskakun@gmail.com
- GitHub: github.com/palinaskakun