I am a Systems-Driven Developer focused on building secure, intelligent, and highly scalable solutions. My expertise lies at the intersection of On-Device Artificial Intelligence (Edge LLMs), Real-Time Cybersecurity, and Scalable Full-Stack Architecture.

• 🎓 Academic Background: B.Tech in Computer Science & Engineering (CGPA: 8.5/10) at ACE Engineering College (2024 - 2028).
• 📈 Proven Code Consistency: Achieved 8,544 contributions in the last year, demonstrating active development cycles and deep technical commitment.
• 🤖 Edge AI Advocate: Architecting lightweight, offline-first systems using local LLMs (Gemma, Qwen) to prioritize data privacy.
• 🛡️ Cybersecurity Researcher: Developing real-time threat-detection engines and neural-shield systems to secure application entry points.
• ⚡ Product Engineer: Bridging the gap between complex research and deployment-ready, highly interactive products.

• Impact: Selected for this national-level open source program. Managed the core architecture of Lexicon, coordinating 794+ commits, 95 branches, 52 pull requests, and 7 contributor forks.
• Engineering: Integrated a Groq AI backend with Llama 3.3, implemented dynamic workspace sharing APIs, and automated 60+ production builds on Vercel.
• Recognition: Ranked #54 on the national NSoC leaderboard.

• Focus: Collaborated on national-level open source repositories. Partnered with mentors and developers across India to build robust front-end interfaces, optimize backend routing, and resolve system bottlenecks.

• Journal: International Journal of Advanced Research in Science, Communication and Technology (IJARSCT) | March 2026
• Core Contribution: Proposed and developed a smart SOS routing architecture that reduces response times by up to 30% by dynamically assigning emergency vehicles to the nearest available hospital/police station based on real-time traffic density and high-priority dispatch systems.
• Links: Research Gate Reference | Sarathi GitHub Repo

Orchestrating local multi-agent systems with zero external API dependencies.

• Multi-Agent Engine: Spearheaded the architecture of a local multi-agent system, orchestrating 3 specialized CrewAI agents (research, writing, execution) to automate task workflows.
• RAG Pipeline: Designed a high-performance Retrieval-Augmented Generation (RAG) system, decreasing document retrieval latency by 45% across 500+ local documents.
• Performance: Successfully processed 100+ daily test queries with a 98% system uptime.
• Tech Stack: CrewAI, Ollama, FastAPI, Next.js, SQLite, Docker

Executing LLMs at the edge with real-time cybersecurity protocols.

• Core Architecture: Native on-device execution of Qwen (0.5B/1.5B) and Gemma (1.8B) models with zero latency or cloud dependencies.
• Neural Shield: Built an asynchronous background analyzer that scans incoming links and traffic in real-time, detecting phishing and malicious inputs before they reach the user.
• Tech Stack: Flutter (Android), Python, Local LLMs (ONNX/Llama.cpp), Secure Storage

Low-latency, privacy-first local assistant.

• Local Speech Pipeline: Built an offline voice interface combining Vosk (STT) and Piper (TTS) with local LLMs, guaranteeing data privacy.
• Performance: Optimized local audio processing to achieve speech-to-text response generation under 200 milliseconds on consumer hardware.
• Tech Stack: Python, Ollama, Vosk, Piper, Local API Server

Client-side document processing and OCR.

• Serverless OCR: Deployed a client-side suite of 10 document tools processing offline text recognition (OCR) across 13 languages.
• Tech Stack: React.js, TypeScript, Tesseract.js, PWA | Live Site

Process monitoring and threat analysis framework.

• System Defense: Designed a high-performance monitor in C++ hooked into local process execution events to flag system anomalies and mitigate threats.
• Tech Stack: C++, Threat Analysis API, Local Process Auditing

A custom dashboard reflecting my actual development highlights and performance benchmarks:

• GitHub Consistency: Logged 8,544 contributions in 2026 alone.
• Competitive Coding: Ranked Top 4 / 200+ in the ACE Code Sprint 2026.
• National Hackathons: Ranked Top 7 in the KL University National Hackathon.
• IIT Bombay: Certified in Python Programming.
• Claude AI Professional: Certified by Anthropic.
• OutSkill: Certified in Generative AI Mastery.

"I believe that code should not only be efficient and secure but should also solve genuine human challenges. The most powerful systems are those that protect user privacy while delivering seamless, high-performance user experiences."