Software Engineer & Cyber Security Undergraduate

Dulith Divisekara

I architect systems and ship them fast: from serverless edge tooling to full-stack booking platforms.

Information Technology undergraduate specializing in full-stack systems, network/edge architecture, and automation. I combine CRUD architecture, MVC design, RESTful thinking, and Linux systems with modern AI-assisted development workflows to design, build, and deploy functional software at a pace traditional workflows can't match.

Software Engineer Automation Engineer Cyber Security Analyst Full-Stack Developer

Prefer the command line? Run my edge-routed resume in your terminal.

curl https://cv.dulithdivisekara.workers.dev
Dulith Divisekara - Software Engineer & Cyber Security Undergraduate

About

Systems Architect, AI-Augmented Builder

I treat AI-assisted development as a force multiplier, not a shortcut. I still design the architecture, define the data model, and own every technical decision; I just execute faster.

My work spans backend systems built from first principles, serverless edge deployment, and agentic infrastructure automation, proof that I can operate across the stack, not just prompt my way through one layer of it.

Formal training in the Model Context Protocol, the Claude API, and agent-skill design means I don't just use AI tools. I understand their architecture well enough to build with and around them.

Academic & Focus Areas

Degree
B.Sc (Hons) in Information Technology
Institution
Sri Lanka Institute of Information Technology (SLIIT)
Target Roles
Software Engineer Automation Engineer Cyber Security Analyst Full-Stack Developer

Skills

Categorized Skills Matrix

Languages, frameworks, systems, and the modern AI-augmented workflows I build with.

Languages & Scripting

Java JavaScript Bash / Shell HTML5 / CSS (Tailwind) Markdown / YAML

Frameworks & Libraries

Spring Boot Thymeleaf Maven Tailwind CSS Cloudflare Workers

Systems, Networking & Security

Linux Fundamentals AWS Compute (EC2, ECS/EKS, Lambda) Serverless Edge Networking Secrets Hygiene Flat-File Data Persistence

Modern Workflow & Architecture

AI-Assisted Prototyping Model Context Protocol Claude API (RAG, Tool Use) Agent Skills Architecture Git / GitHub MVC & CRUD Design Mermaid.js

Work

Featured Projects

Selected builds spanning full-stack systems, serverless edge compute, and agentic AI infrastructure.

Full-Stack Web Application

TixCore Event Booking System

A dependency-light event ticketing platform that gives admins full venue/event management and users a seamless booking flow, without requiring an external database server.

Java Spring Boot Thymeleaf Maven MVC CRUD
  • Modular System Architecture: a 6-module MVC system (Users, Admins, Venues, Events, Bookings, Reviews), each with full CRUD.
  • Custom Persistence Layer: a lightweight, pipe-delimited flat-file data store as a drop-in database replacement.
  • Deployment-Ready Packaging: portable runnable JAR via Maven, single-command deployment with zero external dependencies.
View Source Code ↗

Knowledge Management System

SLIIT Y2S1 Vault

A version-controlled, AI-augmented knowledge base that transforms raw university lecture content across 4 modules into a structured, queryable, retention-optimized learning system.

Markdown Architecture Mermaid.js Git NotebookLM Obsidian
  • Structured Information Architecture: a hierarchical Markdown/frontmatter system compatible with graph-based tools like Obsidian.
  • AI-Augmented Documentation Workflow: LLM-powered study tooling embedding AI-generated quizzes and audio overviews per module.
  • Diagramming-as-Code: dynamically rendered Mermaid.js diagrams plus interactive flashcards for active recall.
View Source Code ↗

Infrastructure Automation

Private Repo

Hermes Backup

A backup and restore pipeline. Kept in a private repository since it handles personal data, but happy to walk through the architecture on request.

Details on request
Private Repository

Interactive CLI / Serverless Tooling

Terminal CV

A programmatic, terminal-native resume served via serverless edge compute, replacing the static personal-website CV with a curl-executable developer experience.

Cloudflare Workers Serverless / Edge JavaScript HTTP Header Parsing
  • Edge-Native Deployment: resume delivery as a serverless function on Cloudflare's global edge network, with low-latency responses.
  • Dynamic Content Negotiation: request-header inspection to detect the client's platform and serve a tailored output format.
  • Differentiated Technical Branding: a curl-executable resume endpoint that signals fluency to a technical audience.
View Source Code ↗

Agentic System State Management

Private Repo

OpenClaw Backup

An autonomous disaster-recovery pipeline for a personal AI agent, enabling full identity, skill-tree, and memory reconstruction on a fresh machine with a single command.

Bash System Automation AI Agent Architecture Security Hardening
  • One-Command Restoration Pipeline: a single script (restore.sh) that reconstructs an entire agent runtime after OS wipe or hardware failure.
  • Security-Conscious Backup Design: credentials stripped from the persisted state, enforcing "restore, then re-inject secrets."
  • Layered State Architecture: isolated directories for identity, capability, and history, transferable to DevOps and IaC practices.
Private Repository

Frontend Web Development

Personal Portfolio (Source Architecture)

A fully responsive single-page portfolio unifying networking, edge-computing, and embedded IoT project work into one cohesive, framework-light brand presence.

HTML5 Tailwind CSS JavaScript Cloudflare Edge Formspree
  • Framework-Light Responsive UI: semantic HTML5 and utility-first Tailwind CSS with vanilla JavaScript, no build tooling.
  • Integrated Contact Pipeline: a serverless contact-form flow via Formspree, cross-linked with the Terminal CV project.
  • Edge-Hosted Delivery: deployed to Cloudflare's Edge Network for global low-latency delivery.
View Source Code ↗

Why Hire Me

"I treat AI-assisted development as a force multiplier, not a shortcut. I still design the architecture, define the data model, and own every technical decision; I just execute faster."

My work spans backend systems built from first principles (a hand-rolled flat-file persistence layer in TixCore), serverless edge deployment across Terminal CV and this portfolio, and agentic infrastructure automation. Formal training in MCP, the Claude API, and agent-skill design means I don't just use AI tools. I understand their architecture well enough to build with and around them. The result is a developer who ships production-shaped software at a pace traditional workflows can't match, without sacrificing the engineering fundamentals that make that software maintainable.

Continued Learning

Certifications

Ongoing, self-directed upskilling across AI tooling, cloud infrastructure, and networking fundamentals.

Contact

Let's Connect

Recruiters, engineering leads, or collaborators: I'd love to hear from you. Drop a message or reach out directly.