Architecting production-grade full-stack systems and ML pipelines that ship to real users. From IIT Patna research labs to Wipro-hardened enterprise infrastructure — I build things that scale and endure.
I'm an AI-first full-stack engineer at IIT Patna, sitting at the intersection of production-grade web systems and applied machine learning. My work doesn't live in notebooks — it ships to real users.
At Thales DIS Technology, I engineered a Linux PAM authentication module in C — cutting manual verification steps by ~40% in the test environment and automating team workflows to reduce manual effort by 80% with zero regressions.
At Wipro, I hardened enterprise infrastructure: configuring Palo Alto firewalls with AppID/URL filtering, running live ARP-poisoning attacks on Active Directory using Kali Linux, and applying CIS Benchmark controls across lab networks.
My projects live in production. DevTrack serves 50+ active developers; NOTE-MINT runs entirely client-side with sub-second load; my Attrition ML System uses XGBoost + SHAP for real-time explainable risk scoring — Dockerised and deployed.
Each project ships to real users or enterprise environments. Code quality, measurable impact, production-grade architecture.
Coding journal with streak tracking, visual insights & public portfolio sharing.
Production app serving 50+ active developers. Streak analytics, tag-based progress logging, Chart.js learning-pattern dashboards, and public portfolio export — all backed by PostgreSQL + Supabase with real-time sync.
ML system scoring attrition risk per employee with full SHAP explainability.
Probabilistic risk scoring with real-time inference via Flask API. Interactive analytics dashboard visualises SHAP feature importance. Fully Dockerised for one-command deployment — production-ready ML pipeline.
Zero-tracking, session-only notes with instant PDF export — no accounts ever.
Runs 100% in-browser. Tab close clears all data. Sub-second load time, single-click PDF export via PDF.js. TypeScript-first architecture with zero backend, zero storage, zero surveillance.
Every skill below has shipped to production or powered IIT Patna research. No padding — only what I've used.
Advanced ML · Deep Learning · NLP · Optimisation Techniques · Probability & Statistics
Data Structures & Algorithms · DBMS · OS · Computer Networks · OOP
Java · Python · TypeScript · SQL · Bash · PowerShell · C — production-tested across all
Open to SDE roles, AI / full-stack collaborations, and freelance projects. Response within 24 hours.
