BSc (Hons) in Computing Science
Dublin, Ireland · Full-time
- Relevant modules: Networks and Data Communication; Linux System Administration; Databases; Web Technologies
Computing Science Graduate 2027 Dublin, Ireland
Final-year BSc (Hons) at Griffith College Dublin - I build, deploy and run what I write
Available for work
about
Final-year BSc (Hons) Computing Science student at Griffith College Dublin, graduating June 2027. Most of what I can do I learned by building and running things outside coursework: a self-hosted Proxmox cluster behind a WireGuard mesh, with Pi-hole, ZFS and Grafana monitoring; a client site in production that syncs its own availability through a GitHub Actions pipeline; a Rails API I audited and hardened after finding raw user input reaching SQL; and a GPU benchmark that holds a LoRA fine-tune constant while polling NVML for thermal telemetry. Earlier, part-time work in industry: building and hardening client workstations, then debugging over 200 SQL-based regulatory reports and reworking around 400 queries against changing compliance rules in a core banking system. I am looking for a graduate engineering role where I can work on real systems and learn from experienced engineers.
skills
Programming Languages
- Java
- JavaScript
- TypeScript
- Python
- SQL
- HTML
- CSS
Backend & Web
- Node.js
- Express
- Spring Boot
- React
- Next.js
- REST APIs
Databases
- MySQL
- PostgreSQL
- MongoDB
Systems & Networking
- Linux (Ubuntu, Debian, Arch)
- Windows provisioning and hardening
- Bash
- Proxmox VE
- ZFS
- TCP/IP
- DNS
- DHCP
- VLANs
- LAN
- WireGuard
- Tailscale
- Nginx Proxy Manager
- Pi-hole
DevOps & Cloud
- Docker
- Docker Compose
- Kubernetes (K3s)
- AWS
- CI/CD Pipelines
- GitHub Actions
- Git
- Grafana
- Uptime Kuma
Security
- Parameterised SQL and injection prevention
- Rate limiting
- Session and cookie hardening
- bcrypt password hashing
- Input sanitisation
- Helmet security headers
- Content Security Policy
- Windows security baselines
AI & Machine Learning
- PyTorch
- Transformers
- PEFT (LoRA)
- TRL
- NVML telemetry
- pandas
- matplotlib
education
Junior Java Developer (certificate)
Online · Part-time
experience
PC Build & Systems Technician
Moscow, Russia · Part-time
- Built and configured 20+ custom workstations to client specification, selecting compatible components and managing thermal and power budgets per build
- Provisioned and hardened the Windows environments shipped with them: security baselines, user access controls and update policy
- Diagnosed component conflicts and hardware faults through structured stress-testing, resolving them before delivery rather than after
Assistant Systems Analyst
Moscow, Russia · Part-time
- Validated and debugged 200+ SQL-based regulatory reports in a core banking system, reconciling output against source data and the applicable reporting requirements
- Wrote and optimised approximately 400 SQL queries behind those reports, reworking slow joins and aggregations as regulatory reporting rules changed
- Supported the rollout of the Diasoft core banking platform at two banks: configuration, testing and production deployment
projects
Infrastructure & Self-Hosting
Self-Hosted Proxmox Cluster
Proxmox VE host running containerised self-hosted services behind a WireGuard mesh, with Pi-hole, ZFS and Grafana monitoring.
Built to cut monthly subscription costs and keep my own data off corporate cloud services. Repurposed hardware runs Proxmox VE hosting containerised services, with a WireGuard mesh for remote access, Pi-hole for DNS filtering, ZFS for storage and Grafana with Uptime Kuma for monitoring. I built it between finishing the Networks and Data Communication module and starting Linux System Administration, and used it to put both subjects into practice.
- Proxmox VE
- Docker
- WireGuard
- Tailscale
- Pi-hole
- Nginx Proxy Manager
- ZFS
- Grafana
- Uptime Kuma
Web Development
Villa Caterina
Client site for a Lake Como rental, built database-free to keep running costs near zero: availability syncs from Booking.com via GitHub Actions, and the IT/FR/DE pages are generated from the English source.
A holiday-rental site for a villa on Lake Como, built for a client who needed running costs as close to zero as possible. That ruled out a database, so availability is pulled from the property's Booking.com iCal feed by a GitHub Actions job every six hours and committed back to the repository as JSON, and the Italian, French and German pages are generated from the English ones by a Python script. It runs on GitHub Pages with a custom domain, a Formspree contact form and a strict Content Security Policy, and it is where I learned GitHub Actions properly.
- HTML
- CSS
- JavaScript
- Python
- GitHub Actions
- GitHub Pages
- Formspree
Pro.me
Service-booking platform built solo, at concept stage: role-based dashboards and session auth on an Express backend with Helmet, rate limiting and input sanitisation.
A service-booking platform I am building on my own: customers browse providers by category, book appointments and leave reviews, and each role gets its own dashboard. It is still at concept stage and currently generic - the plan is to narrow it to driving instructors. The Express backend uses session auth backed by MongoDB, with Helmet, rate limiting, input sanitisation and bcrypt password hashing.
- React 19
- Vite
- Tailwind CSS
- React Router v7
- Node.js
- Express 5
- MongoDB
- Mongoose
- Helmet
- Cloudinary
Sorted
Smartphone catalogue built with a coursemate on a Rails 8 API and a React 19 front end, with server-side filtering across around twenty spec dimensions. I worked on the backend and hardened it: parameterised the raw-SQL filters, added rate limiting, and fixed the session cookie and error handling.
A smartphone catalogue for the Irish market, built with a coursemate: a React 19 and TypeScript front end over a Rails 8 JSON API on PostgreSQL. Filtering runs server-side across around twenty dimensions - brand, screen size, RAM, storage, cameras, refresh rate, release year, 5G, NFC, charging, SIM type and biometrics - and a second endpoint reports how many devices a pending filter set would leave and which brands are still selectable, so the counts update before anything is applied. Filter state is mirrored into the URL, results are paged in chunks of 50, and up to five phones can be put side by side. My work was on the backend: the storage and release-year filters interpolated user input straight into SQL, so I rewrote them onto bound parameters, then added per-endpoint rate limiting, a production-only secure flag on the session cookie, and JSON error handling so an unknown device id returns a 404 rather than a 500.
- React 19
- TypeScript
- TanStack Query
- Zustand
- Tailwind CSS
- Ruby on Rails 8
- PostgreSQL (Neon)
- Vercel
- Render
AI / Machine Learning
AI-GPU-Benchmark
Measures the effect of GPU cooling changes: runs a fixed LoRA fine-tune while polling NVML every 250 ms, then plots thermal and power curves.
I deshrouded my RTX 4070 Ti and fitted custom fans, then had no repeatable way to tell whether it actually ran cooler. This tool puts the card under a fixed AI workload - a 150-step LoRA fine-tune of TinyLlama-1.1B in bfloat16 - while a separate thread polls NVML every 250 ms for temperature, hotspot, VRAM, power draw and clock speed. The PyQt6 window shows live telemetry during the run and plots thermal and power curves when it finishes, so two cooling setups can be compared on identical work.
- Python
- PyQt6
- PyTorch
- Transformers
- PEFT
- TRL
- nvidia-ml-py
- pandas
- matplotlib
contact
Open to graduate engineering roles and internships, and to open-source collaboration. Typically responds within 24 hours.
Availability
Graduating June 2027 and eligible for the Third Level Graduate Programme (Stamp 1G): 12 months of full-time work with no employer permit required.
- Emailvl.rai@proton.me
- GitHubgithub.com/PeGoPaz
- LinkedInlinkedin.com/in/pegopaz
- Xx.com/Vi_Raim
- CVDownload CV (PDF)