Shipping styleEnd-to-end product ownership
Loading…
Available for Work
Hello, I'mZeyedh Rhazes
Software Engineer at Velaris· Master's in AI candidate at Curtin.
I build React/Next.js dashboards, Fastify and Spring Boot backends, secure auth, and AI-assisted hiring pipelines — end-to-end in production.
Software Engineer @ VelarisCurrent Role
Master's in AI · CurtinEducation
React · Next.js · Fastify · Spring BootStack Focus
Colombo, Sri LankaBased in
ReactNext.js (App Router)TypeScriptTailwind CSSViteRechartsTanStack QueryNode.jsFastifyJava 17Spring BootSpring SecurityPython FastAPIPrismaDrizzle ORMZodPostgreSQLMongoDBMySQLAWS LambdaCloudWatchSQSBedrockDockerGitHub ActionsVercelVitestReact Testing LibraryMSWPlaywrightJUnitTestcontainersESLintPrettierReactNext.js (App Router)TypeScriptTailwind CSSViteRechartsTanStack QueryNode.jsFastifyJava 17Spring BootSpring SecurityPython FastAPIPrismaDrizzle ORMZodPostgreSQLMongoDBMySQLAWS LambdaCloudWatchSQSBedrockDockerGitHub ActionsVercelVitestReact Testing LibraryMSWPlaywrightJUnitTestcontainersESLintPrettier
About me
An engineer who ships the whole product, not just the front of it.
I work across the stack — designing the UI, wiring the API, hardening auth, deploying the service, and writing the tests that prove it all holds together. I care about the small details that compound into a real product.
I'm Zeyedh Rhazes, a software engineer at Velarisand a Master's in AI candidate at Curtin University. I build production-style web applications — dashboards, AI-assisted hiring pipelines, microservice backends, secure auth systems, and the full-stack glue that ties them together.
Outside of work I write a small-engineer blog, contribute to open-source curriculum at Curtin's Programming Club, and build personal tools like MyPalrto keep my own skills honest. I'm happiest when the system under my hands has real users and real consequences.
Currently studyingMaster's in AI · Curtin University
Tooling I reach for firstNext.js · Fastify · Spring Boot · Postgres
Where I work fromColombo, Sri Lanka — open to remote
What I actually do
Designing and shipping modular React and Next.js interfaces — from reusable component libraries to full dashboard surfaces with role-based views and typed API client layers.
Building RESTful APIs in Node.js Fastify and Java Spring Boot, with Zod/Spring Security validation, request schemas, proper error handling, and typed contracts.
Implementing auth systems from scratch — RBAC and ABAC permission matrices, Argon2 password hashing, JWT sessions, audit trails, and scoped authorization.
Setting up CI/CD pipelines in GitHub Actions, writing integration/E2E tests with Vitest, React Testing Library, MSW, and Playwright, and keeping code coverage honest.
Working with PostgreSQL and MongoDB at production scale — schema design, indexing, query optimisation, and Prisma/Drizzle migrations.
Bridging AI services into real product flows — Python FastAPI workers, NLP scoring, semantic embeddings, and microservice orchestration with Java + Node.
What I work with
The stack behind everything you see here.
Every tool below has shipped to production at least once. Top tools in each category are marked with a star; chip colour reflects how deeply I work with the tool day-to-day.
Frontend
6/7 prod- React
- Next.js (App Router)
- TypeScript
- Tailwind CSS
- Vite
- Recharts
- TanStack Query
Backend & APIs
6/9 prod- Node.js
- Fastify
- Java 17
- Spring Boot
- Spring Security
- Python FastAPI
- Prisma
- Drizzle ORM
- Zod
Data, Cloud & DevOps
6/10 prod- PostgreSQL
- MongoDB
- MySQL
- AWS Lambda
- CloudWatch
- SQS
- Bedrock
- Docker
- GitHub Actions
- Vercel
Quality & Tooling
4/8 prod- Vitest
- React Testing Library
- MSW
- Playwright
- JUnit
- Testcontainers
- ESLint
- Prettier
What I do
Services I take on as a solo or embedded engineer.
A short, honest list. If what you need isn't here, ask anyway — most of my real work has come from projects that didn't fit a template.
Full-Stack Web Applications
From schema and APIs to dashboards — production-shaped, end-to-end.
I build complete web applications with React/Next.js frontends, Fastify or Spring Boot backends, and a typed contract across the boundary. Schemas are shared, auth is consistent, and CI runs tests on every commit.
- Production-ready REST or hybrid APIs with validation, error envelopes, and audit logs.
- Responsive React/Next.js dashboards with reusable component libraries and typed API clients.
- Database schema, migrations, and seed data ready for local development and CI.
- Quality infrastructure: unit tests, E2E, CI workflows, and a Dockerised local stack.
ReactNext.jsTypeScriptFastifySpring BootPostgreSQLPrisma / DrizzleDocker
Auth, RBAC & Security Layers
Argon2 hashing, JWT sessions, role-based and attribute-based access control.
I've shipped authentication, session management, and granular permission systems for support platforms and HR systems. The same permission model is shared between frontend and backend so a route guard and an API guard never disagree.
- Argon2 password hashing and JWT session handling with refresh and rotation flows.
- RBAC roles (e.g. ADMIN / MANAGER / SUPPORT_AGENT) with shared TypeScript permission tables.
- Audit logs for sensitive actions — who did what, when, against which record.
- Validation, rate limiting, and request-level authorization at the API boundary.
Argon2JWTSpring SecurityZodREST middlewareAudit logging
AI-Assisted Workflows
Resume parsers, JD analyzers, semantic matching, and voice/AI companions.
From my SmartHire capstone to ongoing AI companion work, I integrate Python AI services with Java or Node backends. Multi-service pipelines combine exact, fuzzy, and semantic matching — and the integration points are typed, retried, and observable. This portfolio ships one: a chat widget in the bottom-right with a Groq → DeepSeek fallback chain (try the ✦ AI chat link in the footer).
- Python FastAPI microservices for resume parsing, JD analysis, and semantic matching.
- Hybrid similarity scoring (exact + fuzzy + sentence-transformers) with explainable match scores.
- Cross-language integration (Python ↔ Java/Node) via clean REST contracts and shared schemas.
- AI/voice input experiments and companion flows with offline-friendly fallbacks.
PythonFastAPIsentence-transformersRapidFuzzLlamaCloudJava integration
Performance & Modernization
Move legacy stacks forward without breaking what's working.
Whether it's tightening a Vite build pipeline, optimising a Drizzle query plan, or replacing a slow polling pattern with serverless collectors — I profile first, then ship a small, measurable improvement. No rewrite-everything theatrics.
- Bundle and runtime profiling with concrete size or latency wins documented.
- Query optimisation (Drizzle/Prisma) with indexes, batching, and N+1 elimination.
- Polling-to-event migrations (SQS, CloudWatch) with retry and concurrency controls.
- CI/CD hygiene: smaller pipelines, faster tests, and reliable Docker-based local dev.
ViteTanStack QueryRechartsDrizzle ORMAWS LambdaSQSCloudWatchGitHub Actions
Tooling, Testing & CI
Vitest, Playwright, MSW, Docker, GitHub Actions — quality you can prove.
I've wired the test, mock, container, and CI layers for support platforms, HR systems, and dashboards. Tests run fast, mocks are realistic, and CI fails loudly on the things that matter — flake-free.
- Unit and component tests with Vitest + React Testing Library.
- End-to-end coverage with Playwright against a real local stack.
- MSW request mocks for clean, fast component tests that don't need the network.
- GitHub Actions workflows for typecheck, lint, test, build, and E2E — all in parallel.
VitestPlaywrightMSWDockerGitHub ActionsJunit
Have a project that needs this? Let's talk →
Professional Experience
Where I've been building.
Real work on real systems — from SaaS products to enterprise platforms. Full ownership across frontend, backend, deployment, and CI.
Internship
Software Engineering Intern
Velaris
Building full-stack product features across the Integration Dashboard and PeoplePulse platforms with React, TypeScript, Vite, TanStack Query, Fastify/Node.js, PostgreSQL/Drizzle, Docker, and AWS services. Implemented ABAC for the Velaris CSM Platform and Copilot tool access using raw DB queries and a new Python authorization library.
ReactTypeScriptViteTanStack QueryFastifyDrizzle ORMPostgreSQLDockerAWS LambdaCloudWatchSQSPython
Full-time
Software Engineer — Capstone Project
Aeturnum (SmartHire)
Developed SmartHire, an AI-assisted hiring platform connecting recruiter/candidate UI, Spring Boot core APIs, resume parsing, JD extraction, and matching services combining exact, fuzzy, and semantic similarity scoring. Delivered dashboards, applicant tracking, saved jobs, recommendations, and analytics across five microservices.
ReactTypeScriptJava 17Spring BootSpring SecurityMongoDBMySQLPythonFastAPILlamaCloudRapidFuzzsentence-transformers
Volunteer
Software Engineer
Programming Club of Curtin University
Advised the MANIFEST'24 student team on UI/UX direction — layout structure, visual hierarchy, responsive behaviour, and stakeholder-friendly presentation. Led programming awareness workshops and supported junior students in programming, software engineering fundamentals, and problem-solving practices.
UI/UX designStakeholder communicationMentoringWorkshopsDocumentation
What people say
Straight from the people I've worked with.
Recommendations from managers and collaborators — unedited, in their own words.
Udara Wijeratna1st
Senior VP of Engineering & Country Head, Velaris
I had the pleasure of working closely with Zeyedh at Velaris, and he has consistently demonstrated what it means to be a modern, AI-augmented software engineer.
One of his most impressive contributions was leading the development of IQCenter, a platform that has had a significant impact on proactively identifying and managing integration issues across our customer base. Taking an idea from concept to a production-ready solution that delivers real business value is no small achievement, and Zeyedh played a key role in making it happen.
He also showcased his ability to move at exceptional speed by designing and delivering our HR Management System in just a few weeks. Today, the platform is successfully running in production, highlighting not only his technical capability but also his ability to leverage AI effectively to accelerate software delivery without compromising quality.
Beyond his strong full-stack engineering skills, Zeyedh has an excellent eye for frontend design and user experience. He naturally thinks from the user's perspective, creating solutions that are both technically robust and intuitive to use. This combination of engineering excellence and design thinking is a rare quality that makes him stand out.
What I appreciate most is that his strengths go beyond technology. Zeyedh is an excellent communicator, collaborates effectively across teams, takes ownership of his work, and is always open to feedback and continuous improvement. His positive attitude and strong soft skills make him someone that people genuinely enjoy working with.
I have no hesitation in recommending Zeyedh to any organization. He is an outstanding engineer who combines technical excellence, AI-driven productivity, strong product thinking, and exceptional collaboration skills. I am confident he will continue to make a significant impact wherever he goes.
I wish him every success in the next chapter of his career.
Education
Foundations in software engineering, specialisation in AI.
A formal grounding that maps directly to how I build — strong CS fundamentals, software engineering discipline, and now post-graduate research in applied machine learning.
- Master of Science · Artificial Intelligence
Curtin University
- Selected among a cohort studying advanced AI, machine learning, and statistical reasoning applied to real engineering problems.
Selected courseworkAdvanced Artificial Intelligence Research TopicsExplainable Approaches to Machine LearningNeural Networks and Statistical Machine LearningPlanning and Handling Uncertainty in Machine LearningReinforcement LearningSearch and Logic Approaches in Machine LearningComputer Science ProjectSTEM Research Methodologies - Bachelor of Computing · Software Engineering
Curtin University
GPA 3.9 / 4.0
- Best Performance Dean's Award — Dec 2024, Mar 2025, Nov 2025, Apr 2026, Jul 2026
- Vice-Chancellor's Award, Curtin University — Jul 2026
- Capstone Project: SmartHire (Aeturnum) — multi-service AI hiring platform
Selected courseworkObject Oriented Software EngineeringUnix and C ProgrammingDatabase SystemsIntroduction to Software EngineeringDesign and Analysis of AlgorithmsRequirements EngineeringMobile Applications DevelopmentDistributed ComputingCyber SecurityCloud Computing (AWS)Human Computer Interface - Diploma in Information Technology · Information Technology
UTS College
GPA 6.6 / 7.0
- Distinctions or higher in Web Systems, Programming Fundamentals, Applications Programming, Business Requirements Modelling, Database Fundamentals, and Network Fundamentals.
- Engineering Foundational Studies · Engineering Foundation
UTS College
GPA 6.5 / 7.0
- 4 High Distinctions and 4 Distinctions across the foundational engineering program.
Awards & recognition
- Vice-Chancellor's AwardCurtin University · 2026-07
- Best Performance Dean's AwardCurtin University · Dec 2024, Mar 2025, Nov 2025, Apr 2026, Jul 2026
- Academic Excellence Award — Gold MedalUTS College · 2023 and 2024
Industrial projects
A few products I've shipped end-to-end.
MyPalr, ServixOps, and PeoplePulse — real systems with real users, real constraints, and real tradeoffs. Click any card to explore the interactive demo, or see the full list of projects.
CollaborativeIn progress
MyPalr
A personal life-management companion spanning web, backend, and mobile — tasks, habits, goals, finance, journaling, insights, and AI companion interactions.
Collaborated with Udara Wijeratna — Senior VP of Engineering & Country Head, Velaris
ReactReact NativeExpoTypeScriptNode.jsAI workflows
Key Outcome
Demonstrates cross-platform product thinking — Expo/React Native mobile, AI/voice features, and finance automation for Sri Lankan banks.
PortfolioBuilt
ServixOps
An enterprise support operations platform managing tickets, customers, SLA policies, audit logs, and RBAC-driven workflows.
Next.jsReactTypeScriptTailwind CSSFastifyPrisma
Key Outcome
Demonstrates full ownership of a production-shaped support system — from schema and auth to dashboards and CI.
ProfessionalProfessional contribution
PeoplePulse HR System
A production HR platform for leave lifecycle, approvals, balances, holidays, departments, audit history, and Slack/Google Calendar integrations.
ReactTypeScriptViteFastifyDrizzle ORMPostgreSQL
Key Outcome
Proves CRUD security integration, Drizzle query optimisation, multi-role workflows, and CI/CD discipline at Velaris.