Skip to content
View NadiiaBulmak's full-sized avatar
💫
Open to new opportunities
💫
Open to new opportunities

Block or report NadiiaBulmak

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
NadiiaBulmak/README.md

Hi, I'm Nadiia 👋

Full Stack JavaScript Developer | React • Next.js • Node.js • NestJS • TypeScript • PostgreSQL • AI/RAG • Automation

Full Stack JavaScript Developer with 2+ years of commercial experience building web applications, backend services, integrations, and business automation workflows. Experienced in delivering end-to-end solutions across frontend, backend, databases, APIs, and third-party systems.

⭐ Featured Projects

QuizFlow

Full-stack quiz creation and analytics platform

A production-style full-stack application where authenticated users can create, manage, publish, and share quizzes, while participants can complete public quizzes without an account.

Highlights:

  • JWT authentication with HTTP-only cookies
  • Google OAuth 2.0
  • Protected routes and ownership-based authorization
  • Quiz CRUD and publication workflow
  • PostgreSQL relational data model
  • Prisma ORM
  • Server Actions and Route Handlers
  • Zod validation
  • Server-side score calculation
  • Result persistence and analytics
  • Search, filtering, sorting, and pagination
  • Responsive desktop/mobile UI

Tech: Next.js • React • TypeScript • PostgreSQL • Prisma • Tailwind CSS • Zod

🔗 Live Demo

🔗 Repository

CBT Support Agent

AI support agent for CBT psychoeducation and guided self-reflection

A backend-focused AI agent that provides grounded CBT psychoeducation and guided self-reflection using LLM planning, deterministic safety guardrails, tool calling, and a PostgreSQL/pgvector knowledge base.

Highlights:

  • Multi-step AI agent orchestration with a custom NestJS workflow
  • Gemini-powered planning and answer generation
  • Tool calling with input validation, safety checks, and knowledge-base search
  • PostgreSQL + pgvector semantic search
  • RAG-based grounded responses with source attribution
  • Deterministic safety and out-of-scope guardrails before LLM execution
  • Answer validation with retry and fallback mechanisms
  • Structured JSON response contract with intent, confidence, sources, and decision summary
  • Agent execution and tool-call logging for observability
  • Unit tests and HTTP-level end-to-end tests
  • Swagger API documentation

Tech: NestJS • TypeScript • PostgreSQL • pgvector • Prisma • Gemini API • REST • Vitest • Supertest • Docker

🔗 Repository

📫 Let's Connect

LinkedIn

GitHub

Email


⚡ A little more about me

Before moving into software development, I worked in healthcare and developed strong experience in responsibility, problem-solving, communication, and process management.

Now I combine that background with software engineering to build practical, reliable, and user-focused web applications.

Pinned Loading

  1. next_quiz-builder next_quiz-builder Public

    TypeScript

  2. cat_feeder_frontend cat_feeder_frontend Public

    TypeScript

  3. cat_feeder_server cat_feeder_server Public

    TypeScript