Full-stack cybersecurity investigation platform using React, FastAPI, Neo4j, and graph analytics. Unifies SOC, AML, and identity signals into a single graph. AI agents run autonomous investigations and auto-generate FIU-IND Suspicious Transaction Reports, cutting analyst workload from hours to minutes.

ML + Operations Research platform improving fleet utilization to 94.6%. Combines RandomForest demand prediction (F1=0.93) with OR-Tools route optimization, explainable agentic decision workflows, and scenario simulation engine for dynamic routing strategies.

Multi-agent conversational AI system using LangGraph orchestration. Hybrid RAG pipeline (FAISS + BM25) with source-grounded responses, multi-layer memory including episodic, long-term, and semantic recall. 1.2–1.5s latency with safety guardrails and escalation logic for high-risk emotional states.

AI-powered interactive digital museum with a gamified isometric pixel-art environment and multilingual agentic RAG guide. Users explore Indian heritage through personalized tours in 10+ languages. Winner of the Experience India Track at IndiaStack Build for Billions Hackathon at NITK Surathkal.

Cross-platform fest app (Android, iOS, Web) with 200+ users. Built Polaroid-style memory frames, live map event navigation, gamified QR stall interactions, and vision-based photo clustering using FaceNet so attendees could find photos of themselves automatically. Led backend systems and deployments.

Agentic misinformation detection system that identifies and verifies viral claims during global crises. Specialized AI agents monitor social media and news in real time, cross-checking claims against credible sources. Builds dynamic reputation scores for information sources, enabling faster trust decisions at scale.

AI-powered WhatsApp assistant for GDG on Campus NMIMS that helps students discover events, explore technical domains, and learn about recruitment. Intent classification pipeline with Gemini-powered LLM, conversation state management, and engagement analytics. 100+ students interacted during Foobar outreach.

Multilingual AI voice calling system automating payment follow-ups, improving collection efficiency by 80%. Integrated Twilio, Vapi, WebSockets, and FastAPI for real-time conversations. LLM pipelines for intent extraction, sentiment analysis, and call outcome classification. Async backend with scheduling and rate-limited orchestration.