Mishan Shah
I lead engineering at LQ Digital: production backend systems, server infrastructure, and AI integration. Outside client work, I research why current AI architectures fail and write and stream the journey publicly.
I run backend and infrastructure at LQ Digital. On any given week that is API work, server setup, and whatever AI integration a client product needs, usually across a few projects at once.
Most of what I have built is in private repos. It is client work, so there is not much I can show. The projects below are the parts that are mine to share.
Production runs on VPS I manage myself. I would rather own the stack than hand it to a managed cloud and lose sight of what is actually running.
Alongside the client work, I am researching why current AI architectures fail — specifically the gap between statistical pattern matching and genuine world models. I write about it on the blog and stream the learning process publicly on YouTube.
Personal and academic projects. Client work is under NDA.
NEPSE Trend Analysis
↗An LSTM model trained on Nepal Stock Exchange data. Mostly an exercise in seeing where the model falls apart on a thin, illiquid market.
JWE Decryption Demo
↗A small demo where JWE payloads get decrypted in the browser. The private key never leaves the client, so the server only ever sees ciphertext.
Employee Management
↗A few FastAPI services that talk to each other over signed JWTs, wired together with Docker. Built it to get the service-to-service auth right.
LangGraph RAG Chatbot
PrivateA retrieval chatbot running in production on LangGraph, with pgvector handling lookup so answers stay grounded in the right documents.
Researching AI architectures and writing about it publicly.
Happy to talk backend, infra, or research direction. Drop a message or reach out directly.