building things.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Sardar Vallabhbhai National Institute of Technology, Surat
Data Scientist
June 14, 2026 – Present
Prometheus-Adani
May 7, 2026 – Present
1st place winner at the Adani Hackathon for developing an AI-powered intelligence platform for smart energy grid management. Features real-time demand forecasting, anomaly detection, agent-assisted decision support, live grid simulations, and explainable AI insights. Built with Next.js, FastAPI, LightGBM, Ollama, PyTorch, and scalable ML pipelines.
View ProjectQuiplash
March 28, 2026 – Present
Realtime multiplayer party game inspired by Jackbox Quiplash with live voting, room-based gameplay, and WebSocket synchronization. Features AI-assisted answers, dynamic prompt battles, host/player interfaces, and realtime score tracking. Built with Next.js, PartyKit, React, Tailwind CSS, Groq API, and scalable websocket architecture.
View ProjectGoHTTP-CLI
March 27, 2026 – Present
Terminal-based HTTP/HTTPS traffic inspector with live packet capture, TCP stream reassembly, and interactive TUI analytics. Supports HTTP parsing, TLS SNI inspection, searchable traffic logs, and transaction export functionality. Built with Go, gopacket, Bubble Tea, libpcap/Npcap, and high-performance stream processing.
View ProjectDistributed-RAG
March 27, 2026 – Present
Distributed microservice-based RAG platform for multimodal ingestion, semantic retrieval, and scalable AI-powered querying. Features async embedding pipelines, vector search, reranking, remote worker scaling, and multi-provider LLM orchestration. Built with Next.js, FastAPI, PostgreSQL, Redis, Qdrant, Docker, Ollama, Groq, and Gemini APIs.
View ProjectFinBridge
March 15, 2026 – Present
AI-powered offline-first CRM for intelligent field lead capture, audio insights, and automated follow-ups. Features multimodal sentiment analysis, timestamped transcript navigation, and Google Meet scheduling. Built with Next.js, PostgreSQL, Whisper, Gemma 3, and scalable PWA architecture.
View ProjectIndyChat
May 11, 2025 – July 6, 2025
AI-powered civic assistant for Indianapolis residents, visitors, and public services. Features smart city support, complaint management, voice interaction, and RAG-powered responses. Built with Next.js, Firebase, Dify AI, and ElevenLabs.
View ProjectMinesweeper-Training
November 1, 2024 – November 3, 2024
As I am indeed a nerd, I play a lot of minesweeper in my free time. I thought why not incorporate my field of work with my field of play. So this repo is to make a neural network that can learn to play minesweeper, and be MUCH MUCH MUCH better than my performance. (At least I hope)
View ProjectCultural Fit Analysis
The candidate's portfolio showcases a strong passion for AI/ML and software development, with numerous personal projects demonstrating initiative and continuous learning. The projects are diverse, ranging from energy management to civic tech and gaming, indicating adaptability and broad interests. The target role 'Data Scientist' aligns well with the candidate's demonstrated technical skills and project focus on AI/ML, data forecasting, and anomaly detection. The candidate's experience level is listed as 0, but the depth and complexity of personal projects suggest a higher practical skill level, which could be a good fit for a dynamic, innovation-driven environment.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a proactive and innovative approach to problem-solving, often tackling complex, real-world scenarios (e.g., smart energy grid, civic assistance, distributed RAG). The hackathon win suggests strong teamwork and competitive drive. However, without specific psychometric or English test results, a detailed assessment of communication, logical reasoning, stress handling, and team collaboration is not possible.