Shubham Arora

Backend Engineer · GenAI Builder

Open to backend & AI/ML roles

Systems that scale. AI that ships.

Six years building distributed backends at JPMC, Amazon, Blink Health, and Nutanix. Now deep in LLM systems — RAG pipelines, agent architectures, and the hard engineering problems that make AI reliable in production.

JPMC → Amazon → Blink Health → Nutanix · 2020 → 2026

building: Pensieve

View projects

Years shipping

fintech.retail.healthtech.cloud

Projects built

side quests included

Contributions

380+

past year · GitHub

Last commit

–

github.com/hitawall

“Every day is an exam for me. I want to be able to look at myself in the mirror every single day and tell myself that I did everything I could in order to become better, both on and off the court.”

Viktor Axelsen · two-time Olympic badminton champion

Now

What I'm building

buildingsince Jun 2026

Pensieve

A personal RAG system over a self-maintained knowledge base with hybrid (BM25 + vector) retrieval and cross-encoder re-ranking, benchmarking chunking strategies and vector DB backends (ChromaDB, pgvector) for retrieval quality.

Projects

Things I've built

All projects

AI / LLM2026

Recruiter Reply AI

A chrome extension that puts a draft button on your gmail interface to help you answer those repeated recruiter screen questions.

Hobby2026

Trello Board Sync to Pomofocus

A chrome extension to sync Trello board tasks to pomofocus.io

AI / LLM2026

Auto Job Tracker

A web portal that scrapes off the latest jobs from different job boards and sends emails as per set preferences.

Hobby2026

Forty Rules of Love

An interactive digital retelling of Elif Şafak's novel — one rule surfaced per day, built as a meditative reading experience.

AI / LLM2026

LLM Roadmap

A structured 12-week curriculum for engineers moving into LLM systems work — RAG, agents, evals, and production deployment.

Writing

Recent Thoughts

All posts

Jun 29, 2026

How I use a 20 year old engineering paradigm to keep my AI tokens usage optimized.

With each emerging technology, the fundamentals never change.

Jun 4, 2026

Reinforcement Learning - The paradigm with the power to surpass human intelligence

What is reinforcement learning, and how can it be the tool that advances high levels of thinking by models?

Jun 3, 2026

What are tokens in the context of Large Language Models?

How to explain what tokens are to a kid!

Open to opportunities

Let's build something real

Looking for senior backend or AI/ML engineering roles where depth matters. If you're building systems that need to be fast, reliable, and intelligent – let's talk.