Sheesh Mirza · Software Engineer · AI Builder · Creator
Sheesh Mirza.
Building useful things.
Software engineer focused on Software Engineering & System Designing, Artificial Intelligence, and intelligent automation. Exploring startups, problem discovery, and human psychology.
Interests
Things That Keep Me Curious
01
Software Engineering & System Designing
Backend systems, distributed architectures, APIs, reliability, and the engineering decisions that make software dependable.
02
Artificial Intelligence & Machine Learning
LLMs, AI agents, Model Context Protocol, evaluation, automation, and turning probabilistic models into production systems.
03
Automation and Intelligent Systems
Autonomous workflows, intelligent task execution, and eliminating operational friction through software.
04
Entrepreneurship, Startups & Innovation
Finding problems worth solving, testing ideas with real users, building distribution, and scaling products into sustainable businesses.
05
Human Psychology, Behavior & Habits
Why people buy, how habits form, mental models for clearer thinking, and the behavioral drivers behind technology adoption.
Featured Writing
Notes from the work of building.
Essays on technology, business, psychology, and the decisions that make useful products easier to build and trust.
Work
Projects and experiments
Software, prototypes, and open source repositories fetched live from GitHub.
smirza.in
Personal website and digital garden built with Next.js, React, and Tailwind CSS.
MailHost Web Client
Web application interface for the MailHost email service.
MailHost API Service
High-performance backend API and microservice for MailHost.
Local DevOps & Docker Environment
Some handy Docker scripts to run and keep useful tools locally on your machine. Enjoy! 😜 — Sheesh
Ollama LangChain Client
Ollama Adapter is a small Python command-line example for asking a local Ollama chat model questions through LangChain. It is intentionally lightweight: the repository currently provides the model client and configuration layer, not a document-ingestion or retrieval pipeline.
AI Agents for Beginners
18 Lessons to Get Started Building AI Agents
Hands-On LLMs Code
Official code repo for the O'Reilly Book - "Hands-On Large Language Models"
LLM Course & Notebooks
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
Made With ML Applications
Learn how to develop, deploy and iterate on production-grade ML applications.
AI Model Gateway & Wrapper
Go-based generative AI wrapper and integration proxy.
FreeCharge Hackathon Project
FreeCharge Hackathon 2026
Data Structures in JavaScript
Data structures with JavaScript
LeetCode 30 Days of JS
LeetCode 30 Days of JavaScript
OpenCV Vision & Detection
OpenCV Object Detection with Python
Polyglot Systems Programming
Polyglot programming explorations in C++ and systems languages.
Beliefs
A Few Things I Believe
“Test a small version of your idea today before spending months on an unproven plan.”
“Great software removes effort for people. It never shows off complexity.”
“People don't buy technical specs; they buy safety, status, and peace of mind.”
“A fast backend means nothing if the interface is confusing and hard to use.”
“Shipping fast only helps when you listen closely to real user feedback.”
“Every extra feature brings bugs. Only keep what directly solves an important problem.”
“AI doesn't replace clear thinking; it magnifies sloppy thinking at lightning speed.”
“Understand how the customer feels, not just the technical bug they reported.”
“Simple systems are hard to design, but they are the easiest to trust and maintain.”
“Tight constraints create clean code. Big budgets create slow, bloated software.”
“Prove your idea with real users instead of endless theoretical debates.”
“User trust takes months of reliability to build and one bad design trick to destroy.”
“Master the fundamentals of databases, networks, and memory before chasing trends.”
“One focused builder using modern automated tools can outpace a team stuck in meetings.”