Available for internships & remote roles
I'm Devanshu Jain, a Software, Machine Learning, and AI Automation Engineer building production ML systems, LLM agents, and automation pipelines. Dual-degree Computer Science student at NIT Kurukshetra, shipping things that people actually use — from IoT safety wearables to WhatsApp-native health agents.
About
I'm a CS student at NIT Kurukshetra. I like the intersection of systems thinking and product craft — the kind of engineering where correctness, concurrency, and a real user problem all sit at the same table.
Recent work spans AI-enabled IoT for safety, a production PWA serving hundreds of hostel residents, and an event-driven WhatsApp agent for rural healthcare. On the side I train hard on algorithms — LeetCode rating 1720+, 900+ problems, and a 450+ day streak.
Currently open to software engineering internships, remote roles, and campus placement conversations. If any of the work below resonates, I'd love to hear from you.
Selected work
AI-Enabled IoT SOS Emergency System
Wearable SOS system for women's and elderly safety. Flask backend processes concurrent multi-sensor IoT streams, syncs alerts across Twilio + Google Maps in real time, and runs an Isolation-Forest anomaly pipeline behind a caregiver dashboard. Manuscript in preparation, Dept. of CSE, NIT Kurukshetra.
Smart Mess Management PWA
Progressive Web App digitising hostel mess operations — feedback, attendance, billing — for 400+ residents. Node/Express + MongoDB with indexed queries and concurrent request handling, shipped through automated CI/CD to Render and Vercel.
AI-Powered WhatsApp Healthcare Agent
WhatsApp-native health agent for rural, low-literacy users — vaccination reminders, nutrition screening, outbreak alerts. Event-driven n8n workflows handle concurrent sessions, scheduling, conditional routing, and escalation, with multilingual voice-first interaction.
ML Loan Approval Prediction System
End-to-end supervised ML pipeline for loan approval prediction. Trained and compared KNN, Logistic Regression, and Naive Bayes classifiers with full EDA, feature engineering, and model evaluation. Clean data pipeline with robust preprocessing and visualization.
E-commerce Customer Segmentation System
End-to-end unsupervised ML pipeline that segments 2,240 e-commerce customers into four actionable personas. Engineered behavioural features, reduced dimensionality with PCA, selected K via the elbow method, and compared K-Means with Ward hierarchical clustering. Revealed that ~40% of customers drive ~80% of spend, with a tailored marketing strategy for each segment.
Premium Fitness Coaching Website
Client project for a performance coach — a premium, mobile-first website with dual-zone experience (coaching + lab), WhatsApp-first CTA flow for instant lead capture, integrated Poshak nutrition store, snap-scroll carousels, lazy-loaded images, and edge deployment via Cloudflare Workers.
Recognition
2-member team among 600+ teams · Synersis Group, Hyderabad Campus
Master's Union · Top finalists among 2500+ teams
Toolbox
Leadership
Relations & Outreach — organised NIT Kurukshetra's first global alumni meet, connecting 500+ alumni.
Background
B.Tech + M.Tech Dual Degree, Computer Science & Engineering · CGPA 8.96 / 10
Quick answers
Contact
I reply to every genuine message — recruiter, founder, or fellow student. The fastest path is email; résumé and code are always one click away.