Building production-grade AI backends — not demos, not wrappers. Java · Python · Go · LangChain4j · Spring AI · MCP · AWS
📍 California | 🎓 M.S. Computer Science, UIS '25 | 🔍 Open to Founding Eng / Backend / Cloud roles
AI / LLM
Languages & Frameworks
Cloud & Infra
Databases & Messaging
Certifications
Full-stack platform that helps Shopify merchants review inbound order data, approve line items, and create draft products — powered by LLM-driven PDF extraction and real-time AWS infrastructure.
Role: AI Engineer on a founding team of ~3. Owned end-to-end feature delivery across the full stack.
| Stack | Next.js · TypeScript · AWS (DynamoDB, S3, Lambda, Cognito) · Python · OpenAI API · Shopify API · Vercel · JWT · Socket.io |
| Auth | Multi-tenant signup/login/email verification + Google SSO via Cognito, JWT in httpOnly cookies |
| Listings UI | Order review dashboard pulling live data from DynamoDB + S3-signed image URLs |
| Shopify Sync | Python worker that polls approved items → creates Shopify draft products with idempotency tracking |
| PDF Ingestion | Gmail → Google Drive → OpenAI → structured JSON for automated purchase order extraction |
| CI/CD | Vercel pipeline, 50+ deploys, zero downtime |
🔒 Private org — theSuedeAI — available on request
Social music competition app where producers upload tracks, enter audio battles, vote, and climb leaderboards — built across three microservices with a real-time backend.
Role: Full-stack contributor. Owned bracket logic, track submission flows, and React UI from scoping through production.
| Stack | Node.js · Express · MongoDB (Mongoose) · Socket.io · AWS S3 · JWT · Jest/Supertest · Swagger · React |
| Services | bladesedge (backend) · glacierstorm (frontend) · shadowfury (admin) |
| Core | Battle/voting API, real-time chat + notifications via Socket.io, leaderboard cron jobs |
| Testing | Full pyramid: unit · integration · API · regression · E2E via Jest + Supertest |
| Docs | Swagger UI auto-docs at /api-docs |
🔒 Private org — cntndr — available on request
Computer vision pipeline that inspects product images, classifies defects (Good / Minor / Major), draws bounding box overlays, and generates automated QC reports.
- REST API with single-image, batch, and demo inspection modes
- Swagger / ReDoc auto-docs · React + Vite + Tailwind dashboard
- Edge deployment architecture: YOLO v8 / TensorRT / ONNX / Llama 3.2 for offline ops
🔗 codezenithdev/corseco-qc-assessment
CLI that aggregates and visualizes live metrics from EC2, EBS, ELB, and S3 in one tabular view.
# Real-time EC2 metrics with bastion SSH tunneling
know-my-health list-instances --profile prod --bastion-ip 10.0.0.1 --key-path ~/.ssh/key.pem --visualize
# AWS cost explorer
know-my-health list-cost --profile prod --start-date 2025-01-01 --end-date 2025-01-31- Covers CPU · RAM · IO wait · EBS IOPS/throughput · ELB target group health · S3 ACLs · Cost Explorer
- Multi-profile · tag-based filtering ·
--visualizeflag
| Role | Company | Period |
|---|---|---|
| 🤖 AI Engineer Intern | Suede (AI2 Incubator-backed) | Jan–Mar 2026 |
| 🛠 Backend Developer | Digital Accessibility Remediation, UIS | Jun–Dec 2025 |
| 💻 Software Developer | MQuotient Pvt. Ltd. | Jan–May 2024 |
| 🏥 Software Developer | Peakhealth Solutions | Dec 2022–Dec 2023 |
| ⚙️ Full Stack Intern | Controlytics AI | May–Dec 2022 |
🎓 M.S. Computer Science — University of Illinois Springfield (Dec 2025) 🎓 B.Tech Information Technology — Government College of Technology, Coimbatore
"If it works, don't touch it." — until there's a better architecture.