Dewang Gondaliya

Dewang Gondaliya

Full-Stack Developer | Applied AI Systems Engineer

About Me

Full-Stack Developer with 1+ year of experience building production web applications using React, Next.js, Node.js, and PostgreSQL. Currently working on SBI's HRMS Claims Module, contributing to modernization of legacy systems and performance optimization in an enterprise environment serving large user bases.

Hands-on experience designing AI-assisted workflows using LLM APIs, structured context engineering, and interactive UI systems for content generation platforms.

Strong foundation in data structures, clean architecture, and performance debugging. Focused on scalable system design, maintainable abstractions, and production reliability.

Skills

Frontend

React.jsNext.jsMaterial UITailwind CSS

Backend

Fast APINode.jsExpress.js

Databases

MongoDBPostgreSQLMySQLQdrant Vector DBPinecone

AI / LLM Systems

RAG system designPrompt EngineeringContext StructuringLLM API IntegrationAgentic Workflow Design

State & Data Handling

Redux ToolkitAxiosREST API Integration

Programming Languages

JavaScriptTypeScriptPythonCC++

Tools & Practices

GitGitHubClean ArchitectureComponent ReusabilityPerformance Optimization

Work Experience

Webito Infotech

React developer

Dec 2024 - present

SBI HRMS Claims Module

  • Refactored legacy React components to improve maintainability and reduce defect propagation across modules.
  • Implemented code-splitting and lazy loading to optimize bundle size and improve load performance.
  • Restructured component architecture to enforce separation of concerns and reusable UI patterns.
  • Collaborated with backend engineers to integrate new claim types and ensure API contract consistency.
  • Performed deep debugging of production issues across state management and rendering layers.
  • Applied clean architecture principles to reduce tight coupling between UI and business logic.

Kreato.AI (June 25 - Present)

  • AI-powered YouTube content assistant (Thumbnail, Title, Script)
  • Integrated YouTube Data API to fetch channels and videos as input sources for AI workflows.
  • Developed the Title Generation Flow using a ChatGPT-style UI for interactive title suggestions.
  • Built the Script Generation Flow with a Canvas-based interactive UI: Allowed users to add YouTube links, documents, and audio files. Enabled creation of input groups to structure contextual data for AI processing.
  • Collaborated closely with AI Integration Team to define: Structured context input formats from the frontend. Expected response handling patterns on the backend.
  • Conducted R&D on context design and prompt structuring to ensure AI responses were relevant and user-specific.
  • Set up the project architecture and created reusable UI components to enable scalable development with minimal code duplication.

Projects

AI Content Generation Engine (Workflow Automation)

n8n, LLM APIs, Webhooks, Structured JSON Output

Designed and deployed an automated content generation pipeline using n8n as workflow orchestrator.

Engineering Signal

Workflow orchestration, LLM output validation, structured prompt design, production automation.

Book-Referenced AI Agent (Local LLM with Retrieval)

Ollama, Open-source LLM, Embeddings, Vector Store, Python

Built a local AI agent capable of answering questions grounded in book content using open-source LLMs via Ollama.

Engineering Signal

RAG system design, embedding pipelines, local inference trade-offs, hallucination mitigation.

Education

Bachelor in Computer Application

Shri Atmanand Saraswati Science College, VNSGU

CGPA - 7.64

2022 - 2025

Certifications

Gen AI and Agentic workflows
Prompt Design in Vertex AI Skill Badge
Introduction to Artificial Intelligence and Machine Learning
Figma Bootcamp

Get In Touch

I'm currently looking for new opportunities. Whether you have a question or just want to say hi, I'll try my best to get back to you!