Computer Science Engineer.

leetcode

github

Abinash Ojha  • Peerlist

Comprehensive Cloud-Based Practice Management & Nutrient Analysis Software for Ayurvedic Dietitians, Tailored for Ayurveda-Focused Diet Plans.” 🌿 The Challenge: Ayurvedic dietitians currently lack a digital-first tool to manage patients and generate diet plans rooted in Ayurvedic principles. Most platforms today focus only on modern nutrition, leaving a gap for holistic, Ayurveda-driven solutions. ✨ Our Solution: AyurDiet – A Cloud-Based Ayurvedic Diet Software 🔹 Blends Ayurveda (Dosha, Rasa, Virya, Vipaka) with modern nutrition (calories, protein, carbs, etc.) 🔹 Auto-generates personalized, doctor-ready 7-day diet charts in minutes 🔹 AI/ML-powered suggestions for smarter, personalized recommendations 🔹 Structured, searchable, and shareable digital replacement for handwritten charts 🔹 Scales across solo dietitians, clinics, and hospitals ⚙ Tech Stack & Flow: Frontend: React + Tailwind CSS Backend: Spring Boot, Spring WebFlux, Spring Security, JWT Auth, RBAC (Admin/Doctor/Patient) Database: postgresql Model Integration: Spring WebFlux for real-time ML requests ML Models: TensorFlow,Pandas 🚀 Deployment: Frontend → Netlify Backend → GitHub → Render Database → neon.tech server less postgress

Ayur-Diet

🔗 Clixora – URL Shortener with Analytics Spring Boot, React, MySQL, Redis, JWT, Docker, Distributed Deployment Clixora is a scalable full-stack URL shortening platform designed with performance optimization, analytics readiness, and secure API architecture. It enables users to generate short links using random or custom aliases while tracking engagement through click analytics and visual insights. I engineered the backend using Spring Boot and optimized redirect performance with Redis caching, reducing average redirection latency by ~70–80% and significantly improving response time under repeated access patterns. Redis-based fixed-window rate limiting was implemented to protect APIs from abuse and control traffic bursts, improving reliability by ~30–40% during high request scenarios. The system uses JWT-based stateless authentication via Spring Security to secure user workflows and link ownership. A modular REST API architecture supports future analytics expansion (click tracking, geo insights, device metrics). The backend is containerized with Docker and deployed on Render, with MySQL on Neon and a React + Tailwind frontend on Netlify, demonstrating distributed cloud deployment and service isolation. This project strengthened my skills in backend performance engineering, caching strategy design, rate limiting, and production-grade full-stack deployment. Key Highlights: Redis caching • Rate limiting • JWT security • Analytics-ready schema • Dockerized deployment

Clixora

🏫 GIETUVerse – Real-Time Virtual Campus Metaverse (Spring Boot, WebSockets (STOMP), React, PostgreSQL) GIETUVerse is a real-time 2D virtual campus simulation of GIET University that enables users to explore the environment as live avatars and communicate through proximity-based private chat. I built this project to move beyond theory and gain practical understanding of WebSocket protocols, real-time synchronization, and event-driven system design. The system uses Spring Boot WebSockets with the STOMP messaging protocol to establish persistent bidirectional communication between clients and server. Each connected user appears as a live entity on the campus map, with movement controlled via keyboard input. When a user moves, updated coordinates are sent to the backend and instantly broadcast to all subscribed clients, ensuring synchronized spatial awareness across sessions. The platform also supports proximity-based private messaging. When two users come within a defined spatial radius (~50 units), the server enables direct messaging channels. User sessions are uniquely mapped via email identifiers to route private messages to the correct WebSocket connection, demonstrating targeted messaging and real-time presence tracking patterns used in multiplayer and collaborative systems. The frontend was built in React to render live positional updates, while the backend is deployed on Render and the frontend on Netlify with Neon serverless PostgreSQL. This distributed deployment mirrors modern cloud architectures and supports scalable real-time interaction. Key Highlights: Real-time state sync • Broadcast & private messaging • Spatial chat logic • Event-driven architecture

GIETUVerse – Real-Time Virtual Campus Metaver

⚖️ Legal Document Analyzer – Optimized LLM + RAG Backend Spring Boot, Gemini LLM, RAG, PgVector, Redis, JWT I rebuilt and optimized my Legal Document Analyzer into a production-oriented LLM system that performs clause extraction, summarization, risk detection, and document-specific Q&A using Retrieval-Augmented Generation (RAG). I redesigned the LLM response pipeline to produce structured outputs via prompt templates and parsed them directly on the backend into a relational schema mapped to legal headings. This eliminated large JSON parsing on the frontend, reduced payload size, and improved query performance and UI rendering efficiency. To enable document-specific querying, I implemented a full RAG pipeline using embeddings stored in PostgreSQL with PgVector. Documents are chunked and embedded; user queries are converted to vectors and matched via similarity search to retrieve top-K relevant context before LLM inference. This reduced hallucinations and significantly lowered token usage and cost while improving answer accuracy. Redis-based fixed-window rate limiting protects the system and controls LLM cost (2 analysis requests/day/user; 5 standard API calls/hour). The backend uses JWT stateless authentication and modular AI service integration. This rebuild provided hands-on experience with real-world RAG pipelines, vector search, LLM cost control, and scalable AI backend architecture. Key Highlights: RAG + PgVector • Structured LLM outputs • Redis rate limiting • Cost-aware LLM design