Spandan Joshi

Spandan Joshi

Full-Stack Developer · AI & Web3

About

I'm a passionate full-stack developer with a focus on building impactful tools and intelligent systems. I work across modern technologies like Generative AI, LLMs, Langchain, React, Web3, Node.js, and MongoDB to craft seamless, high-performance applications.

I combine strong UI/UX design sense with deep backend and database expertise—ranging from SQL, NoSQL, and Graph, to Vector databases. My interests lie in AI-driven solutions like Retrieval-Augmented Generation (RAG), Agents, and model fine-tuning, and I thrive on solving complex problems while contributing to the tech community.

Work Experience

Xneeti Tech Private Limited Gurugram, India

Software Engineer

Mar 2025 – Present

  • • Built scalable backend services using NestJS, PostgreSQL, RabbitMQ, and BullMQ
  • • Developed ETL pipelines, RBAC system, and analytics reports for internal and client use
  • • Focused on performance, reliability, and modular service architecture
Clapingo Remote

Founding Engineer

Sep 2020 – Mar 2025

  • • Led architecture, backend (Node.js, Express, MongoDB), and frontend (React) development
  • • Managed AWS/DigitalOcean infrastructure, CI/CD pipelines, and deployments
  • • Built and scaled web apps for learners, teachers, and admins
  • • Hired and mentored developers to grow the product team

Skills

JavaScript TypeScript React Node.js Python PostgreSQL AWS Docker MongoDB Express NestJS RabbitMQ BullMQ Airflow Redis Terraform

Projects

LUMIN – Intelligent Data Analysis Platform
GitHub | Demo

Developed a data analysis platform enabling users to connect databases and query data in natural language with dynamic visualizations.

  • • Integrated LangChain framework and multi-LLM support (Groq, OpenAI, Hugging Face) for enhanced analytical capabilities
  • • Built backend with FastAPI, PostgreSQL, and Docker, frontend with React.js and Tailwind CSS
AI Real-time Voice Agent
GitHub

Built a real-time voice-to-voice AI system with streaming pipeline, using STT, LLM, and TTS for natural and seamless conversations.

  • • Integrated STT (Deepgram) for transcription, LLMs (OpenAI, Groq) for context-aware responses, and TTS for fluent, natural speech synthesis
  • • Developed a robust backend with FastAPI and WebSockets, Redis for queue, and a React.js frontend for low-latency, scalable operations
  • • Achieved a significant latency reduction from 30 seconds to 2 seconds

Education

GIET University Gunupur

Engineer's degree, Electronics and Communication Engineering

2016 - 2021

Saraswati +2 Science College

High school, Science

2014 - 2016