Technical Expertise Breakdown
Programming Languages
C, C++, Java, Python, SQL, JavaScript, TypeScript, Rust
Web Technologies & Frameworks
HTML, CSS, JavaScript, TypeScript, React.js, Next.js, Tailwind CSS, Node.js, Flask, Express.js
AI & Machine Learning
LangChain, OpenAI API, Hugging Face Transformers, FinBERT, pgvector, Pinecone, XGBoost, Gemini API, Sarvam TTS/STT
Databases & Infrastructure
PostgreSQL, MongoDB, Supabase, Redis, Upstash
Systems & DevOps
Rust (Axum, Tokio), Git, GitHub, Linux, WebSockets, WebGPU, Docker, Vercel, CI/CD
Projects by Hrushi Bhanvadiya
RasoiAI — AI-Powered Restaurant Management Platform
Problem: Restaurants struggle with static pricing and manual order processing, missing out on data-driven revenue optimization.
Solution: Architected a full-stack multi-tenant SaaS platform featuring an XGBoost ML pipeline for dynamic pricing and an end-to-end multilingual AI voice ordering module using Sarvam TTS/STT and the Gemini API.
Architecture: React frontend with Node.js REST API backend. FastAPI microservice for ML inference. PostgreSQL for transactional data. XGBoost model trained on historical sales data for price optimization. Voice pipeline: Sarvam STT for speech-to-text in 22+ Indian languages, Gemini API for intent extraction, Sarvam TTS for response generation.
Tech Stack: React, Node.js, FastAPI, PostgreSQL, XGBoost, Gemini API, Sarvam TTS/STT
Impact: Automated phone-based order placement across 22+ languages and surfaced per-item pricing recommendations via REST API.
GitHub Repository
Axon OS — Browser-Based Operating System
Problem: Traditional portfolios are static pages that fail to demonstrate real systems engineering capability.
Solution: Built a fully browser-based OS with draggable windows, a virtual file system, and an AI copilot powered by Gemini. Rust backend (Axum + Tokio) streams real-time system metrics via WebSockets.
Architecture: Next.js frontend with custom window manager, virtual file system implemented in TypeScript, and real-time system metrics streamed from a Rust backend via WebSockets. AI copilot uses Gemini API for natural language system control. WebGPU used for hardware-accelerated rendering.
Tech Stack: Next.js, TypeScript, Rust (Axum, Tokio), WebGPU, WebSockets, Gemini API
Impact: Real-time OS simulation running entirely in-browser with sub-100ms WebSocket latency.
GitHub Repository
FinGuide — Neurodivergent-First Fintech Platform
Problem: Existing fintech apps overwhelm neurodivergent users with complex interfaces, increasing financial anxiety.
Solution: Created an adaptive fintech app with 3 UI complexity modes (Default, Simplify, Visual) that dynamically adjust based on cognitive stress detection. Integrated Gemini-powered AI financial assistant and visual budget forecasting.
Architecture: React frontend with adaptive UI state machine supporting ADHD, dyslexia, and anxiety profiles. Node.js backend with TypeScript. Gemini API integration for conversational financial advice. Visual budget forecasting with interactive charts. Accessibility-first component library built with Tailwind CSS.
Tech Stack: React, TypeScript, Node.js, Tailwind CSS, Gemini AI
Impact: Accessibility-first design supporting ADHD, dyslexia, and anxiety profiles with adaptive UI switching.
GitHub Repository
Financial Document RAG System
Problem: Searching through large-scale financial documents for specific answers is slow and error-prone for analysts.
Solution: Engineered a retrieval-augmented generation pipeline using FinBERT embeddings with Pinecone vector indexing. Responses are grounded with inline citation references to source documents.
Architecture: Python Flask backend with document ingestion pipeline. FinBERT for domain-specific financial text embeddings. Pinecone vector database for similarity search. LLM-powered answer generation with citation grounding. Chunking strategy optimized for financial document structure (tables, footnotes, headers).
Tech Stack: Python, Flask, Pinecone, FinBERT, LangChain
Impact: Citation-grounded Q&A over 10,000+ financial documents with sub-2-second retrieval latency.
GitHub Repository
DB Index Visualizer
Problem: Students and engineers lack intuitive tools to understand how database indexing structures operate internally.
Solution: Built interactive HTML5 Canvas visualizations for LSM Trees, Bloom Filters, and Skip Lists with step-by-step operation tracing and real-time state rendering.
Tech Stack: JavaScript, HTML5 Canvas
Impact: Step-by-step visual debugging of insert/search/delete across 3 data structures.
GitHub Repository
CPU Scheduling Simulator
Problem: OS scheduling algorithms are difficult to grasp without visual, interactive simulation.
Solution: Simulated FCFS, SJF, SRTN, and HRRN scheduling with support for dynamic process arrivals, preemption logic, and automatic Gantt chart generation.
Tech Stack: JavaScript, Flask, Tailwind CSS
Impact: Full scheduling simulation with visual Gantt charts and performance metric comparisons.
GitHub Repository
Clinical AI — Medical Document RAG
Problem: Medical professionals need reliable, citation-backed answers from verified documents, not hallucinated AI responses.
Solution: Built a context-aware medical query app using RAG to ground responses exclusively in verified medical documents, with side-by-side source viewing and citation verification.
Tech Stack: JavaScript, Python, CSS, HTML
Impact: RAG-grounded medical Q&A with inline document citation verification.
Internship Scraper & Matcher
Problem: Manually searching for relevant internships across multiple job boards is time-consuming and imprecise.
Solution: Automated scraper extracting software engineering internships from LinkedIn and Glassdoor, with an intelligent matching algorithm that filters results against parsed resume preferences.
Tech Stack: Python, Web Scraping, Data Processing
Impact: Automated extraction and preference-matched filtering across 2 major job boards.
LAPD Crime Analysis — Data Visualization Platform
Problem: Raw LAPD crime datasets are massive and difficult to interpret without visual analysis tools.
Solution: Built a data analysis and visualization platform processing LAPD crime statistics into interactive charts, heatmaps, and statistical summaries with filtering capabilities.
Tech Stack: HTML, JavaScript, Python
Impact: Interactive crime data visualization deployed on Vercel with real-time filtering.
GitHub Repository | Live Demo
Frequently Asked Questions about Hrushi Bhanvadiya
Who is Hrushi Bhanvadiya?
Hrushi Bhanvadiya is a full-stack software engineer and AI developer from Ahmedabad, India. He is a Computer Science undergraduate at the Institute of Technology, Nirma University (ITNU). He specializes in building high-performance web applications, RAG systems, and agentic AI workflows using React, Next.js, TypeScript, Python, and Rust.
What technologies does Hrushi Bhanvadiya use?
Hrushi works with React, Next.js, TypeScript, Python, C++, Rust, Node.js, Flask, PostgreSQL, MongoDB, Supabase, LangChain, OpenAI API, Hugging Face, Pinecone, pgvector, and Tailwind CSS. His primary stack is TypeScript/React/Next.js for frontend, Node.js/Python for backend, and Python with LangChain for AI applications.
What projects has Hrushi Bhanvadiya built?
Hrushi has engineered 9+ complex systems including RasoiAI (AI-powered restaurant SaaS with multilingual voice ordering), Axon OS (browser-based operating system with Rust backend), FinGuide (neurodivergent-first fintech platform), Financial Document RAG (citation-grounded Q&A over 10K+ documents), Clinical AI (medical document RAG), DB Index Visualizer, CPU Scheduling Simulator, and LAPD Crime Analysis platform.
What is Hrushi Bhanvadiya's expertise in AI?
Hrushi specializes in Retrieval-Augmented Generation (RAG) systems, agentic AI workflows, and LLM application development. He has built production RAG pipelines using FinBERT embeddings, Pinecone vector databases, and LangChain. He also has experience with XGBoost for ML-based pricing optimization and Gemini API for conversational AI.
Where does Hrushi Bhanvadiya study?
Hrushi is pursuing a B.Tech. in Computer Science and Engineering at the Institute of Technology, Nirma University (ITNU) in Ahmedabad, Gujarat, India. His current CGPA is 8.72/10.
Is Hrushi Bhanvadiya available for hire or internships?
Yes, Hrushi is actively seeking software engineering internships and selective freelance opportunities. He is open to roles in full-stack development, AI/ML engineering, and systems architecture. Contact him at hrushibhanvadiya@gmail.com.
What kind of systems does Hrushi Bhanvadiya build?
Hrushi builds high-performance web applications, AI-integrated platforms, real-time systems with WebSockets, browser-based operating systems, multi-tenant SaaS platforms, RAG pipelines for document Q&A, and data visualization tools. His focus is on scalable architecture, type safety, and measurable performance.
How can I contact Hrushi Bhanvadiya?
Contact Hrushi via email at hrushibhanvadiya@gmail.com, on GitHub at github.com/hrushi2501, or on LinkedIn at linkedin.com/in/hrushi-bhanvadiya-081818280.
What is Hrushi Bhanvadiya's competitive programming rating?
Hrushi has a LeetCode rating of 1798 and a Codeforces rating of 1217, demonstrating strong algorithmic problem-solving skills in areas like dynamic programming, graph algorithms, and advanced data structures.
What is RasoiAI?
RasoiAI is an AI-powered restaurant management SaaS platform built by Hrushi Bhanvadiya. It features an XGBoost ML pipeline for dynamic menu pricing and a multilingual AI voice ordering module supporting 22+ Indian languages using Sarvam TTS/STT and Gemini API.
What is Axon OS?
Axon OS is a browser-based operating system built by Hrushi Bhanvadiya using Next.js, TypeScript, and Rust. It features draggable window management, a virtual file system, an AI copilot powered by Gemini, and real-time system metrics streamed from a Rust (Axum + Tokio) backend via WebSockets.
What is Hrushi Bhanvadiya's approach to software engineering?
Hrushi follows an architecture-first methodology: deep problem analysis, clear system boundary design, incremental implementation with continuous testing, performance optimization using profiling and Lighthouse audits, and deployment with CI/CD pipelines. He prioritizes type safety (TypeScript, Rust), modular architecture, and measurable performance metrics.
Also Known As
Hrushi Bhanvadia, Rushi Bhanvadiya, Hrush Bhanvadiya, Hrooshi Bhanvadiya, Hrushy Bhanvadiya, Hrushee Bhanvadiya, Hrishi Bhanvadiya, Krushi Bhanvadiya, Khushi Bhanvadiya, Hrushi Bhanvadya, Hrushi Banvadiya, Hrushi Bhanwadiya, Hrushi Bhanvaidya, Hrushi Bhanvadi, Rushi Bhanvadia, Hrushy Bhanvadia, Hrushii Bhanvadiya, Hrushi2501, hrushi2501, Hrushi 2501.