Software Engineer specializing in backend systems, AI integration, and scalable full-stack architectures. I build production-grade platforms with real users, real scale, and measurable impact — systems serving 30K+ monthly users and sustaining 1000+ concurrent traffic.
- ⚡ Software Engineering — Production-focused, performance-driven system design
- 🧠 AI/ML Systems — On-device LLM inference, transcription pipelines, intelligent automation
- 🌐 Full-Stack Development — Next.js, React, Node.js, FastAPI, PostgreSQL
- 📦 Product Engineering Mindset — Shipping measurable outcomes: 56.7% latency reduction, 95% faster processing, 20% efficiency gains
- 🏆 Finalist @ TensorCraft ML Hackathon (NIT Trichy)
🔓 Open To: Software Engineering Roles · AI Engineering Roles · Backend Engineering · Open Source Collaboration.
🔓 Fun Fact: I believe coders are good mathematician.
| Domain | Proficiency | Details |
|---|---|---|
| Edge AI / On-Device Inference | ████████░░ | Local LLM inference with WebAssembly + WebGPU, privacy-first architecture |
| Speech & Audio AI | ████████░░ | Faster Whisper transcription, segmentation, caching — 95% processing speedup |
| Machine Learning | ███████░░░ | Scikit-learn pipelines, predictive modeling, resource optimization |
| AI-Driven Systems | ████████░░ | Real-time disaster prediction, misinformation filtering, inventory intelligence |
| AI Integration | █████████░ | Production AI APIs, FastAPI services, preprocessing optimization |
💼 Crack Off-Campus — Career & Jobs Platform
Production career platform with 30K+ monthly users, rated ⭐ 4.9 with 1K+ downloads and 260+ reviews.
| Attribute | Details |
|---|---|
| Stack | Next.js, React.js, Node.js, PostgreSQL |
| Scale | 30K+ monthly users, 1K+ downloads |
| Performance | Backend latency reduced 30ms → 13ms (56.7%) |
| Security | JWT Authentication, RBAC, secure Razorpay payments |
| Impact | ⭐ 4.9 rating across 260+ reviews |
| Repository | github.com/thelegendaryarticuno |
Engineered a full-stack career platform with optimized backend query paths, role-based access control, and integrated payment infrastructure — built for sustained production traffic with measurable latency improvements.
🎪 siNUsoid Platform — Tech Fest Management System
Live event platform sustaining 1000+ concurrent users with real-time hackathon dashboards.
| Attribute | Details |
|---|---|
| Stack | React, Node.js, Socket.IO, CMS |
| Scale | 1000+ concurrent users on live platform |
| Performance | Real-time dashboards with low-latency Socket.IO streams |
| Security | Role-scoped admin CMS access |
| Impact | 20% admin efficiency gain for data extraction |
| Repository | github.com/thelegendaryarticuno |
Architected and deployed a real-time fest management system across Vercel, Azure VM & Cloudflare R2, powering live hackathon operations with instant data dashboards and streamlined admin workflows.
🎙️ AI Transcription System — Podcast Intelligence Pipeline
AI pipeline converting podcast audio into topic segments in 45–60 seconds.
| Attribute | Details |
|---|---|
| Stack | FastAPI, Faster Whisper, React JS |
| Scale | Repeated-upload optimized processing pipeline |
| Performance | 45s → 2s (95% faster) for repeated audio uploads |
| Security | Isolated processing with validated upload handling |
| Impact | Near-instant repeat transcription via intelligent caching |
| Repository | github.com/thelegendaryarticuno |
Built an end-to-end transcription and topic segmentation system with preprocessing optimization, smart segmentation, and caching layers — turning a 45-second AI workload into a 2-second response for repeat content.
📄 DocuMentor — Privacy-First AI Document Assistant
Fully on-device AI document assistant — zero backend AI calls.
| Attribute | Details |
|---|---|
| Stack | React, TypeScript, WebAssembly, WebGPU, Local AI Runtime |
| Scale | Browser-native — scales with zero server cost |
| Performance | Local LLM inference accelerated via WebGPU |
| Security | 100% on-device processing — documents never leave the browser |
| Impact | Smart Highlights, Student Mode, Research Mode for drafting & debate |
| Repository | docu-mentor-woad.vercel.app |
A privacy-first Edge AI system running local LLM inference entirely in-browser — featuring Smart Highlights for key topics, a Student Mode for guided learning, and a Research Mode for drafting and debating over documents.
🚨 ResqAI — AI Disaster Resource Optimization
AI engine predicting real-time disaster resource needs.
| Attribute | Details |
|---|---|
| Stack | Python, ML Models, Real-Time Data Pipelines |
| Scale | Multi-zone disaster severity mapping |
| Performance | Real-time prediction & resource allocation |
| Security | Validated survivor-tracking data flows |
| Impact | AI-driven inventory management for emergency response |
| Repository | github.com/thelegendaryarticuno |
Built an AI system for emergency operations — combining severity mapping, survivor tracking, and predictive inventory management to optimize disaster resource distribution in real time.
📡 DisasterPulse — Real-Time Disaster Intelligence
Aggregated disaster intelligence from social media + news with AI misinformation filtering.
| Attribute | Details |
|---|---|
| Stack | Python, AI/NLP Filtering, Geo-Mapping, Dashboards |
| Scale | Multi-source real-time data aggregation |
| Performance | Live geo alerts + trend prediction |
| Security | AI-verified information pipeline |
| Impact | Solves disaster data chaos with verified intelligence |
| Repository | github.com/thelegendaryarticuno |
A real-time disaster intelligence platform that aggregates noisy multi-source data, filters misinformation with AI, and surfaces geo-targeted alerts, dashboards, and predictive trends.
🆘 QuickAid — Emergency Assistance App
Mobile-first emergency communication — alerting nearby people instantly.
| Attribute | Details |
|---|---|
| Stack | Mobile (APK Prototype), Real-Time Alerts |
| Scale | Proximity-based alert broadcasting |
| Performance | Instant nearby-user notification |
| Security | Location-scoped alert delivery |
| Impact | Real-world rapid emergency response system |
| Repository | github.com/thelegendaryarticuno |
A rapid-response emergency app prototype enabling instant, proximity-based alerts — designed for mobile-first communication when seconds matter.
2023 — Present
Engineered and operated the official tech fest platform serving live event traffic at scale.
- Sustained 1000+ concurrent users on production infrastructure
- Built real-time hackathon dashboards with Socket.IO
- Improved admin data-extraction efficiency by 20% via custom CMS
- Deployed across Vercel, Azure VM & Cloudflare R2
React Node.js Socket.IO Azure Cloudflare R2 Vercel
2023 — Present
Built and scaled a career platform from zero to 30K+ monthly users.
- Reduced backend latency 30ms → 13ms (56.7%)
- Implemented JWT Auth, RBAC & Razorpay payment integration
- Shipped full-stack features across Next.js, React, Node.js, PostgreSQL
- Achieved ⭐ 4.9 rating with 1K+ downloads & 260+ reviews
Next.js React Node.js PostgreSQL JWT Razorpay
| 🏅 Recognition | Details |
|---|---|
| TensorCraft ML Hackathon Finalist | Finalist at NIT Trichy's machine learning hackathon |
| Production Scale | Built systems handling 30K+ monthly users & 1000+ concurrent traffic |
| Performance Engineering | 56.7% backend latency reduction · 95% AI processing speedup |
| Product Impact | ⭐ 4.9 rated platform with 1K+ downloads & 260+ reviews |
| Consistency | 1200+ GitHub contributions per year |
learning:
- Distributed Systems & System Design
- LLM Inference Optimization (WebGPU / WASM)
- Advanced Backend Performance Engineering
building:
- Privacy-first Edge AI tooling
- Scalable production platforms with real users
- AI-driven real-time intelligence systems
exploring:
- On-device AI runtimes
- Event-driven architectures
- Cloud-native deployment patterns
open_to:
- Software Engineering Roles
- AI / ML Engineering Roles
- Open Source Collaboration




