Computer Science and Engineering student at ULAB in Dhaka. I build Arduino and ESP32 systems, automation tools and self-hosted web servers, and I use AI to work faster, smarter and more efficiently.
I am a B.Sc. student in Computer Science and Engineering at the University of Liberal Arts Bangladesh (ULAB). I was born in Lalmonirhat district, in Bangladesh's Rangpur Division, and I now live in Dhaka. I learn by building: a radar prototype, a wireless vehicle, a field sales app, a WordPress site running from a phone.
Since 2020 I have worked in IT support, computer and electronics project management, and app development for schools and colleges. More recently I moved into field sales, which taught me how real businesses track orders, routes and stock. That is why much of my software work targets practical, low-cost tools.
I speak Bangla (native) and English (intermediate, professional working level), and I can read and write Arabic. I am also comfortable reading and writing mathematical and scientific notation, formulas and equations. I have taken part in debate competitions and presented academic and technical projects at fairs and exhibitions.
A public WordPress site served from an Android smartphone, with Cloudflare Tunnel providing HTTPS access. Includes automated startup and recovery, monitoring, and encrypted backups to Google Drive.
Termux, Nginx, PHP-FPM, MariaDB, WordPress, Cloudflare Tunnel, Bash, rclone
A radar-style monitoring prototype built around an Arduino and ultrasonic sensors with supporting hardware. A Java desktop application provides the graphical interface for viewing what the sensors detect.
Arduino, ultrasonic sensors, Java (GUI)
A point-to-point controller and receiver pair that drives a terrain vehicle without needing a Wi-Fi network, using an analog joystick and dual-motor driver.
ESP32, ESP-NOW, L298N, Arduino C++
A mobile tool for logging outlet orders and daily route collections. It converts cartons to pieces automatically and feeds daily and monthly reports.
AppSheet, Google Sheets, Google Apps Script
Multi-OS network print server using CUPS and LPD over port 515. Dual-boot setups with GPT/EFI partitioning and boot-entry management through efibootmgr. Android debloating with Shizuku and GSI.
Linux, CUPS, UEFI, Termux
An 8S 24V LiFePO4 pack with a smart BMS and buck converters, built to keep small electronics running during outages.
LiFePO4, JK BMS, XL4016
I use AI to work faster, smarter and more efficiently. It handles the repetitive and exploratory parts, while I set the goals, make the decisions and check every result.
Manus, Cursor and Windsurf for multi-step work such as building, refactoring and research, with an agent doing the legwork while I direct and review.
Claude Code and Gemini CLI working directly in the terminal with project files, scripts and servers.
Google Workspace Flows, n8n and Notion to connect forms, sheets, documents and notifications into routines that run on their own.
Connecting AI assistants to tools and data sources through the Model Context Protocol, so they can act on real information.
ChatGPT, Gemini, Claude, DeepSeek and Meta AI for research, debugging, drafting and explanations, choosing the model that fits the task.
NotebookLM for source-based research, and batch processing of documents and images for repetitive jobs.
I test what AI produces on real hardware and real servers before I rely on it.