I am a Robotics and Artificial Intelligence Engineer currently pursuing an M.S. in Robotics & Autonomous Systems with an Artificial Intelligence concentration at Arizona State University. I hold a B.Tech in Artificial Intelligence & Machine Learning from SVKM's NMIMS MPSTME and have hands-on experience across AI, ML, Deep Learning, Computer Vision, NLP, Data Analytics, Robotics, XR, and Intelligent Automation.
My work focuses on building practical, scalable, and intelligent systems that solve real-world problems. Through academic projects, internships, and independent development, I have worked on applications including healthcare AI, visual recognition, NLP-based classification, predictive analytics, robotics systems, XR testing, IoT systems, and AI-driven automation.
I am especially interested in creating intelligent autonomous systems that combine perception, decision-making, and human-centered design to improve productivity and real-world interaction.
| Project | Summary | Tech Focus |
|---|---|---|
| ASU Robotics AB Site | Built a published robotics lab website and documentation interface for ASU Design School Robotics Lab, presenting KUKA KR10/KRC5 details, workflow references, safety resources, and robotic fabrication knowledge-base navigation. | Robotics Documentation, Next.js, HTML |
| RAS-545 Robotics Systems 1 | Developed robotics-focused lab work involving system-level modeling, robotics concepts, algorithmic problem solving, and simulation-oriented implementation. | Robotics, Python, Jupyter Notebook |
| EEE-598 Deep Learning Foundations & Applications | Implemented advanced deep learning labs covering neural networks, attention mechanisms, multi-head attention, GANs, diffusion models, GNNs, and accelerator-based workflows. | Deep Learning, Python, Jupyter Notebook |
| Visual Question Answering Model | Built an AI pipeline designed to answer questions from visual inputs by combining image understanding with intelligent reasoning. | Computer Vision, Deep Learning, AI |
| AI-Based Noise Cancellation Model | Developed an intelligent noise reduction model focused on improving audio clarity through AI-based signal processing. | AI, ML, Audio Processing |
| Project | Summary | Tech Focus |
|---|---|---|
| Satellite Imagery or Drone Feed Change Tracker | Built a geospatial intelligence dashboard and computer vision pipeline to compare satellite imagery or drone feeds, detect land/water/urban changes, and export map-ready change artifacts. | Geospatial CV, React, TypeScript, OpenCV |
| X-Ray Fracture Detection — 92.5% Accuracy | Built a fracture classification workflow for X-ray scans, including preprocessing, model training, evaluation, and performance visualization. | Healthcare AI, Deep Learning, Python |
| 140k Real vs Fake Detection — 97.395% Accuracy | Developed an image classification model to distinguish real and fake images using a large-scale dataset and deep learning workflow. | CNN, Image Classification, Deep Learning |
| Image Classification Using CNN | Implemented a CNN-based image classification pipeline for visual pattern recognition and feature learning. | CNN, Computer Vision, Python |
| Face Recognition Using Haar Cascade | Created a face recognition system using Haar Cascade-based detection and Python scripts for dataset creation and recognition. | OpenCV, Python, Face Recognition |
| Leaf Disease Detection System | Built a deep learning image classification system to identify plant leaf diseases and support agricultural decision-making. | Deep Learning, Computer Vision |
| Real-Time Fruit Detection | Developed a real-time computer vision pipeline for fruit detection and object recognition from visual input. | OpenCV, Object Detection, Python |
| Project | Summary | Tech Focus |
|---|---|---|
| Nebius Chat | Built a polished Streamlit AI studio for Nebius Token Factory models with multi-model chat, image generation, custom instructions, reasoning visibility, export tools, token tracking, and usage analytics. | Streamlit, Nebius, Image Generation, LLM Apps |
| RouteLLM Chat | Developed an intelligent chat application that automatically routes prompts between a cost-effective Nebius Llama model and GPT-4o-mini using RouteLLM, with Streamlit and hosted Next.js implementations. | RouteLLM, Streamlit, Next.js, LiteLLM |
| Talk to Your DB | Built an AI-powered MySQL query assistant that converts plain-English business questions into safe read-only SQL using Nebius Token Factory, schema-aware prompting, result explanations, CSV export, and Streamlit/Next.js interfaces. | Nebius, MySQL, Streamlit, Next.js |
| Calendar Assistant | Built an Agno-powered scheduling agent that connects Nebius AI with Cal.com for availability checks, booking creation, rescheduling, cancellation, upcoming booking lookup, CLI usage, timezone conversion, and a deployable Next.js project website. | Agno, Cal.com API, Nebius, Next.js |
| Smart Scheduler Assistant | Developed an Agno multi-agent productivity assistant that reads Gmail, filters important scheduling emails, coordinates email and calendar agents, manages Google Calendar events, stores action memory in SQLite, and includes a polished Next.js website. | Agno, Gmail API, Google Calendar, SQLite |
| Web Automation Agent | Built a Browser Use web automation agent powered by Nebius Token Factory that follows natural-language instructions to control a real browser, search ecommerce pages, extract product data, and return results in Markdown. | Browser Use, Nebius, Python, Web Automation |
| Newsletter Generator | Built an AI-powered newsletter generator that researches topics with Firecrawl, orchestrates generation through Agno, uses Nebius AI for writing, and provides a Streamlit interface with markdown download and deployment-ready files. | Streamlit, Agno, Firecrawl, Nebius |
| Reasoning Agent | Developed a transparent financial reasoning agent that breaks investment questions into assumptions, reasoning steps, tool calls, risk profiling, allocation suggestions, and balanced recommendations. | Agno, Nebius, Financial Reasoning |
| Agno UI Example | Built an end-to-end Agno Agent UI playground with Nebius-powered web research and finance agents, persistent SQLite conversations, DuckDuckGo search, YFinance tools, and a Next.js chat interface. | Agno, Next.js, YFinance, DuckDuckGo |
| Mastra Weather Bot | Created a Mastra AI weather agent that uses Nebius AI, Open-Meteo weather data, TypeScript tools, Zod schemas, a CLI bot, and a Vite/GitHub Pages website for live weather lookup. | Mastra, TypeScript, Open-Meteo, Nebius |
| KAOS Starter | Built a Kubernetes-native KAOS multi-agent starter with coordinator-worker agents, MCP tool servers, agent memory endpoints, hosted Ollama/SmolLM2 model runtime, deployment scripts, and a published project website. | KAOS, Kubernetes, MCP, Ollama |
| Finance Agent | Developed an Agno-powered AI finance research assistant that combines Nebius-hosted DeepSeek, Yahoo Finance market data, and DuckDuckGo search for stock analysis, fundamentals, analyst recommendations, and market news. | Agno, Finance AI, YFinance, Nebius |
| Human-in-the-Loop Agent | Built a Python human-in-the-loop AI agent that pauses tool execution for approval, rejection, or retry, demonstrating safer agent workflows with Agno pre-execution hooks, Rich terminal UI, and a React/Vite web demo. | HITL Agents, Agno, Python, React |
| Agno HackerNews Analysis Agent | Built an Agno-powered Hacker News analysis agent that studies stories, comments, engagement signals, trending themes, and discussion density using Nebius Qwen models. | Agno, Nebius, Qwen, Hacker News Tools |
| OpenAI SDK Starter | Created an OpenAI Agents SDK starter using Nebius Token Factory and Resend email tools to demonstrate practical agent workflows with external tool integrations. | OpenAI Agents SDK, Nebius, Resend |
| LlamaIndex Task Manager | Built a LlamaIndex ReAct agent starter for task duration analysis, estimation, and productivity-focused reasoning using Nebius-powered inference. | LlamaIndex, ReAct Agent, Nebius |
| CrewAI Research Crew | Developed a CrewAI research crew powered by Nebius AI to coordinate emerging technology research, analysis, and structured report generation. | CrewAI, Research Agents, Nebius |
| Letta Starter | Built a Letta starter demonstrating persistent agent memory blocks, stateful conversations, Docker-based Letta server setup, and Nebius Token Factory inference. | Letta, Persistent Memory, Docker, Nebius |
| PydanticAI Weather Bot | Created a PydanticAI weather assistant powered by Nebius AI and DuckDuckGo search for structured tool-using responses and live information retrieval. | PydanticAI, Nebius, DuckDuckGo Search |
| LangChain Starter | Built a LangChain starter project focused on creating a simple LLM agent workflow with reusable structure for prompt, model, and tool experimentation. | LangChain, LLM Apps, Agent Starter |
| LangGraph Starter | Developed a minimal LangGraph ReAct agent starter powered by Nebius Token Factory for graph-based agent reasoning and tool execution workflows. | LangGraph, ReAct Agent, Nebius |
| AWS Strands Agent Starter | Built a Strands Agents weather assistant using Nebius Token Factory, LiteLLM, and the National Weather Service API to demonstrate live API tool calling. | Strands Agents, LiteLLM, Weather API |
| Mastra Starter | Created a minimal TypeScript Mastra starter powered by Nebius Token Factory for experimenting with TypeScript-based AI agent development. | Mastra, TypeScript, Nebius |
| Camel AI Starter | Built a Camel AI starter project to explore multi-agent collaboration patterns and agent workflow experimentation. | Camel AI, Multi-Agent Systems |
| DSPy Starter | Developed a Nebius-powered DSPy ReAct starter agent with Wikipedia search, Python computation, and a published project website. | DSPy, ReAct, Wikipedia Search, Python |
| Google ADK Starter | Built a Google ADK email-enabled agent using Nebius AI Studio and Resend to send structured HTML emails through a practical tool-calling workflow. | Google ADK, Nebius, Resend, Email Agents |
| Semantic Kernel Starter | Created a Microsoft Semantic Kernel ChatCompletionAgent starter powered by Nebius Token Factory for structured chat-agent experimentation. | Semantic Kernel, ChatCompletionAgent, Nebius |
| Smolagents Starter | Built a Smolagents starter project to explore lightweight agent development patterns and tool-using AI workflows. | Smolagents, AI Agents, Tool Use |
| AutoGen Starter | Developed a Microsoft AutoGen starter powered by Nebius Token Factory for experimenting with conversational and multi-agent AI workflows. | AutoGen, Multi-Agent AI, Nebius |
| cagent Starter | Built a Nebius-powered cagent multi-agent technical writing workflow with a coordinator, web search agent, and writer agent configured through YAML. | cagent, Docker Agent, Multi-Agent Writing |
| Sayna Voice Agent | Created a realtime Sayna voice agent starter with STT, TTS, WebSocket streaming, REST APIs, LiveKit helpers, and a published website. | Voice Agents, STT, TTS, WebSockets |
| Multi-Agent Orchestration System | Built a Python-based multi-agent orchestration framework with hierarchical agent handoffs, recursive task delegation, internal tools, MCP support, cycle detection, and configurable execution depth. | Python, AI Agents, MCP, Tool Orchestration |
| Multi-Modal E-Commerce Fake Review Detector | Built a multi-modal fake review detection system that combines review text, reviewer metadata, and product image signals with an explainable FastAPI scoring pipeline and deployable ML workflow. | PyTorch, Transformers, FastAPI, Docker |
| Real-Time MLOps Drifting & Monitoring System | Developed a real-time MLOps monitoring app that simulates production drift, tracks model metrics, raises alerts, manages retraining history, and stores runtime state using Firebase. | MLOps, React, Firebase, Drift Monitoring |
| Codec Technologies Projects | Built applied AI projects including stock price prediction using Linear Regression and Twitter sentiment classification using NLP and Logistic Regression. | ML, NLP, scikit-learn, Python |
| Basic ML Projects | Worked on practical machine learning notebooks including video game sales prediction and USA accidents data analysis. | ML, Data Analytics, Python |
| GTZAN Genre Classification | Built a music genre classification workflow using the GTZAN dataset for audio-based pattern recognition. | Audio ML, Classification, Python |
| Statistical Methods | Practiced statistical analysis and foundational data science methods through notebook-based implementations. | Statistics, Data Science |
| Spam Email Detection | Created an NLP-based spam classification model for intelligent email filtering using text preprocessing and ML classification. | NLP, ML, Python |
| Video Game Sales Prediction | Built a predictive analytics model to analyze and forecast video game sales trends from structured data. | Predictive Modeling, Data Analytics |
| Project | Summary | Tech Focus |
|---|---|---|
| Autonomous CI/CD Failure Patcher | Built a Firebase-backed self-healing CI/CD dashboard and GitHub Action that analyzes failed workflow logs, generates AI-assisted patches, verifies fixes in an isolated emulator workflow, and opens pull requests. | GitHub Actions, Firebase, React, TypeScript |
| Enterprise MCP Database Server | Developed a secure Firebase-backed web app and MCP SDK server for database metadata workflows, including connector fingerprints, schema snapshots, privacy-aware SQL generation, execution-plan inspection, anomaly explanation, and Codex MCP client export. | MCP, Firebase, React, TypeScript |
| AI-Assisted Robotic Design Interface for Creative Fabrication | Built a futuristic research-interface prototype that maps creative design intent into robotic fabrication workflow parameters with human-in-the-loop review and readiness scoring. | React, TypeScript, Robotic Fabrication |
| Digital Fabrication Curriculum and Workshop Documentation Toolkit | Created a curriculum support toolkit for robotic digital fabrication workshops, including modules, experiment outlines, safety checklists, diagrams, templates, and assessment rubrics. | Documentation, Fabrication, Workshop Design |
| Job Applications Tracker | Built a job application tracking platform to organize applications, monitor progress, manage statuses, and streamline the job search workflow. | Web Development, Productivity, Application Tracking |
| B2B Competitor Intelligence Engine | Developed an intelligence-focused platform for tracking competitors, analyzing business insights, and supporting B2B decision-making workflows. | B2B Intelligence, Analytics, Web Development |
| Service Provider Platform | Developed a JavaScript-based service provider platform focused on connecting users with services through a structured application workflow. | JavaScript, Web Development |
| MyIdea | Built a JavaScript project focused on product-style implementation, front-end logic, and idea prototyping. | JavaScript, Web Development |
| Secure Online Banking System | Designed and implemented a secure DBMS-based banking system with structured data handling and transaction workflows. | DBMS, SQL, System Design |
| IoT Gesture-Controlled Bluetooth Speaker | Built an IoT-based Bluetooth speaker controlled through gesture-based interaction for hands-free operation. | IoT, Embedded Systems, Sensors |