Software Engineer · Backend · Cloud · AI

Vardan Pal

ALT +000.4

THR 000%

Scroll ToExplore

01 // About

Full stack, fewer bottlenecks

I work across the backend and cloud to build fast, scalable, and secure architectures.

I have experience working with Node.js, FastAPI, PostgreSQL, MongoDB, Docker, and AWS, shipping security-first Corrective RAG systems, autonomous PR reviewers, and AI-driven routing engines.

I focus on building production-ready architectures—optimizing queries, enforcing strict security guardrails against prompt-injections, and deploying via automated CI/CD pipelines.

▹ B.Tech, IT — NIT Srinagar (2022–2026)
▹ AWS Certified Cloud Practitioner
▹ AWS AI Practitioner (AIF-C01) Prep Course
▹ JPMorgan Chase & Co. SWE Simulation
▹ Anthropic Claude Course Certificates

02 // Where I've worked

Software Engineering Intern – Cloud Native Applications @ Noventiq (Umbrella Infotech)

Nov 2024 — Feb 2025 · Noida, India

Optimized MongoDB queries, refactored RESTful APIs, and containerized Node.js environments.

▹Optimized MongoDB compound indexes across 12 microservices, reducing p95 query latency from 800ms to 240ms.
▹Refactored RESTful APIs using modular Node.js/Express.js; boosted Jest test coverage from 40% to 75%.
▹Containerized Node.js environments via Docker Compose, reducing developer onboarding from 2 days to 30 mins.
▹Implemented GitHub Actions CI/CD pipelines supporting code reviews, agile releases, and production deploys.

// Systems check

Skill modules online

FLY THROUGH THE CALIBRATION CORRIDOR

03 // Some things I've built

Projects in orbit

▸ CLICK A CARD TO INSPECT

Explore the archive ↗

04 // What's next

Let's make something together.

I'm currently open to new opportunities — full-time, contract, or just a good chat. Whether you have a project idea, a question, or you just want to say hi, my inbox is the best way to reach me.

vardan2701@gmail.com →

Opens your mail app — I reply within 24h

GitHub · LinkedIn · Medium

Initializing Launch Sequence

About

I work across the backend and cloud to build fast, scalable, and secure architectures.

I have experience working with Node.js, FastAPI, PostgreSQL, MongoDB, Docker, and AWS, shipping security-first Corrective RAG systems, autonomous PR reviewers, and AI-driven routing engines.

I focus on building production-ready architectures—optimizing queries, enforcing strict security guardrails against prompt-injections, and deploying via automated CI/CD pipelines.

Based in Marathahalli, Bangalore.

Experience

Software Engineering Intern – Cloud Native Applications — Noventiq (Umbrella Infotech)

Nov 2024 — Feb 2025 · Noida, India

Optimized MongoDB queries, refactored RESTful APIs, and containerized Node.js environments.

Optimized MongoDB compound indexes across 12 microservices, reducing p95 query latency from 800ms to 240ms.
Refactored RESTful APIs using modular Node.js/Express.js; boosted Jest test coverage from 40% to 75%.
Containerized Node.js environments via Docker Compose, reducing developer onboarding from 2 days to 30 mins.
Implemented GitHub Actions CI/CD pipelines supporting code reviews, agile releases, and production deploys.

Projects

DocGuard

June 2026 – July 2026 · Security-First Corrective RAG Agent — Corrective RAG with LangGraph, pgvector, and Claude

Developed a Corrective RAG agent using LangGraph for semantic retrieval, query rewriting, and grounded generation with Claude relevance grading. Engineered security layers against prompt injection, PII, and secret leaks; deployed via FastAPI/MCP with Redis caching and Langfuse observability.

Built with: Python, LangGraph, FastAPI, pgvector, Claude, MCP, Docker.

DevAgent-X

Feb 2026 – Apr 2026 · Autonomous PR Reviewer & Sandbox — Claude 3.5 code reviews with Docker sandbox

Built an automated PR review pipeline integrating Express.js webhooks and Claude 3.5 Sonnet for real-time code analysis and feedback loops. Secured untrusted code execution within isolated Docker sandboxes, implementing strict guardrails against prompt injection and secret leakage risks.

Built with: Node.js, TypeScript, Express.js, Claude 3.5, Docker, Webhooks.

LogiPulse

Jan 2026 – May 2026 · AI-Driven Supply Chain Routing — Live routing with Dijkstra/A* and RAG

Designed Dijkstra/A* routing algorithms for 500+ node paths, integrating live weather and fuel APIs for dynamic supply chain route planning. Created a GPT-4o OCR and RAG pipeline for structured manifest extraction, leveraging PostgreSQL and MongoDB for efficient workload separation.

Built with: Next.js, TypeScript, PostgreSQL, MongoDB, GPT-4o, Prisma.

Skills

Backend: Node.js · Express.js · FastAPI · REST APIs

Databases: PostgreSQL · pgvector · MongoDB · Redis · Prisma

Cloud & DevOps: AWS · Docker · GitHub Actions · CI/CD · MCP

AI & LLM: LangGraph · RAG · Vector Search · Prompt Eng

Security: Prompt Injection Def · PII Detection · Sandboxing

Languages: Python · TypeScript · JavaScript · SQL · C++

Vardan PalVardan Pal