Full-Stack Engineer · Systems Architect · Software Craftsman
Building scalable systems, engineering intelligent tools, and exploring the boundaries of technology.
I'm Rakibul H. Rabbi — a software engineer passionate about enterprise application development, system architecture, developer tooling, automation, and intelligent software systems.
Currently working at Ha-Meem Group, where I contribute to enterprise ERP systems, production workflows, HRM integrations, Windows-based services, and business-critical applications.
My engineering experience spans .NET, C#, SQL Server, Vue.js, React, Angular, Laravel, and modern JavaScript ecosystems.
I enjoy solving problems that emerge when systems become complex. Rather than repeatedly solving the same engineering challenges, I prefer building reusable tools, abstractions, and scalable foundations that make development easier for everyone.
Beyond professional development, I experiment with local AI infrastructure, LLM-powered applications, developer productivity tools, and automation.
“Good engineering is not about writing more code. It is about making complexity understandable.”
Eight companies and ten roles since 2019, from WordPress and Laravel sites to the enterprise ERP behind Ha-Meem Group’s HRM, Payroll and Production.
Enterprise ERP: HRM, Payroll and Production.
Dhaka, on-site
“The Business That Makes Sense”
Full e-commerce site with complete WooCommerce functionality.
The tools I reach for, grouped the way the work uses them. No percentages: either I ship with it or it isn’t listed. Pick a stack to see the tools one piece of work connects.
The local AI workspace: a Vue client over .NET services, SQL Server for state, Ollama for the models.
C# and .NET for enterprise services; Laravel and Node.js where a project already lives.
Vue and Nuxt by preference; React, Angular and Next.js when the team calls for them.
SQL Server every working day: schema design and the queries production depends on.
IIS and Windows services in production; Docker, Nginx and Linux for newer work.
Ollama, local models and context engineering: the ground Mostisko stands on.
RabbitIC (মস্তিষ্ক, “the brain”) is an extensible, locally hosted AI workspace that brings together conversation, context engineering, document intelligence, automation and enterprise workflows. The goal is a local AI ecosystem that helps developers, reads enterprise documents, works with structured data and supports business operations. It is in active development; planned modules are drawn with broken lines.
OddityIn development
An API gateway built around Ollama-compatible requests and responses, adding contextual processing, request tracking, analytics and per-user configuration.
PIPILIKA first, since this page borrows its look. Then work and client sites, one experiment written up as a case study, and one made for the fun of it. Every picture is a screenshot of the live site.
Attendance Management Software
“Because every second matters.”
Open PIPILIKA (opens in a new tab)This page takes its colours, type and cards from it.
“Fabrics are mostly made in Bangladesh; accessories were not. We make them here: pocketing, interlinings, labels.”
Optimizes CVs for applicant tracking systems.
“Stop Writing Code. Start Controlling Value.”
Read: Launching Calcifer 2.0 (opens in a new tab)Art and sign-symbol gallery.
একসাথে আড্ডা, সিনেমা এবং বই পড়া — এখন এক জায়গাতেই।
Hangouts, movies and reading together, all in one place.
Election Data Entry Module
A Vue 3 and TypeScript admin panel where election officers enter, search and filter vote records.
A Vue 3 + TypeScript admin panel where election officers enter, search and filter vote records. It is one module of a larger system with Election Summary, Users, Common Data, Operations and Election configuration sections.
String theory and the mathematical black hole.
Six courses from Anthropic, completed in 2026: the Claude API, the Model Context Protocol, Claude Code and Cowork, and AI fluency.
Credential ID 72z8cpbdgivn
Credential ID 9msurd5d76bp
Credential ID jpoab7kzusx2
Credential ID bb57c6dnfbwf
Credential ID j4mer56opqqf
Credential ID 58w5kxx7asvs
A working notebook of what I test beyond the day job: what each experiment set out to do, where it stands, and what it has shown so far.
Ongoing
Run useful language models on company hardware so documents never leave the building.
Ongoing
Give a small local model the right few pages instead of everything, and measure the difference.
Ongoing
Keep the Ollama request shape so existing clients work, and add tracking, analytics and per-user settings behind it.
Running in production
Cut the overhead of reporting screens that only read data in ASP.NET Core applications.
Running in production
Stop rebuilding auth, RBAC and project structure for every new .NET service.
Notes from the work: .NET performance, databases, local models and the habits that keep systems readable. Published on Medium, with older notes kept on this site.
Have you ever noticed that an AI model can give different answers to the same question? Sometimes the response is direct and predictable.
A .NET 8 Microservice Template for Teams That Want Enterprise Basics Without Enterprise Complexity.
When building ASP.NET Core applications, it’s common to have multiple database contexts.
Let’s engineer a solution.
Opens your email app with this as the first line, addressed to me.
