Full-Stack SDE | Node.js · React/Angular · MongoDB · ClickHouse · Kafka | Building real-time data pipelines & backend systems @ SupplyNote

Noida

Kanpur




Bachelor of Engineering - BE, Computer Science & Engineering
Dr. Bhim Rao Ambedkar University, Agra
Join Ayush and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossFull-Stack SDE | Node.js · React/Angular · MongoDB · ClickHouse · Kafka | Building real-time data pipelines & backend systems @ SupplyNote
I'm a full-stack software engineer working on supply chain and payments infrastructure for the F&B industry at SupplyNote. My day-to-day spans backend APIs, data pipelines, and frontend — Node.js, AngularJS, MongoDB, ClickHouse, and Kafka. I care about systems that are fast, reliable, and built to scale. Some things I've shipped: → AutoGRN system using Python + LLaMA OCR on GCP — cut invoice entry time by 90% → Kafka + ClickHouse analytics pipeline over 10M+ records — report gen from 90 min to under 10 min → Debezium CDC connector streaming 100K+ transactions/15 min from MongoDB to Kafka → Refactored MongoDB aggregation pipelines — API latency dropped from 120s to under 5s on 4M+ records → Razorpay payment integration with webhook-based reconciliation → Sentry across full stack — 2× reduction in system error rate I also enjoy DSA and have solved 750+ problems on LeetCode and GFG. Open to connect with engineers working on interesting backend and data problems.
Noida
claude code
claude
github copilot
Software Engineer at SupplyNote, Sep 2024 - Present · 2 yrs, Noida, • Built autoGRN system (Python + LLaMA OCR, GCP Cloud Functions) to extract structured line-items from invoice PDFs — reduced GRN entry time by 90% for bulk processing. • Co-architected Kafka + ClickHouse real-time analytics pipeline over 10M+ MongoDB records — cut report generation from 90 min to under 10 min. • Set up Debezium CDC connector streaming 100K+ transactions/15 min from MongoDB to Kafka; built async worker process pool for guaranteed downstream delivery. • Engineered PO vs GRN reporting with ClickHouse + Parquet-based batch ingestion — 40× faster than the original MongoDB JSON pipeline (40 min → 1 min). • Refactored MongoDB aggregation pipelines with cursor-based pagination — eliminated socket timeouts, API latency down from 120s to under 5s on 4M+ records. • Integrated Razorpay payment gateway — server-side order creation, signature verification, webhook handling for real-time payment reconciliation. • Integrated Sentry across AngularJS + Node.js stack — error tracking, distributed tracing, 2× reduction in system error rate.
Software Engineer Intern at Hoverin Aerospace, Jan 2024 - Jun 2024 · 6 mos, Kanpur, • Led backend development with Node.js, Express.js (TypeScript), MongoDB, and PostgreSQL (Prisma ORM) — shipped auth, weather data, and drone telemetry APIs. • Built internal operations dashboard with React, Next.js, and Tailwind CSS. • Deployed and maintained services on AWS EC2 with CI/CD pipelines for staging and production.
Bachelor of Engineering - BE, Computer Science & Engineering, Dr. Bhim Rao Ambedkar University, Agra
github: https://github.com/soulmini