• Hello, I'm Rafidul

    With 4 years of experience as a software engineer and a really good eye for detail, I can help you build reliable systems end-to-end. You can check out my work here.


    My Work Download Resume
    Next ->

  • About Me

    As a full-stack developer with approximately 4 years of experience, I love designing and optimizing systems that can solve real world problems. Working with startups and different clients, I've gained experience working with TypeScript, Python and Go to build scalable backend systems and responsive UI in Next.js.

    Looking ahead, I'm enthusiastic about broadening my horizons by delving into the fascinating world of emerging AI technologies to integrate them into new applications in actual helpful ways for the users.

    Currently, I'm sharpening my system design skills to better understand distributed architectures, scalability, reliability, and performance optimization for modern applications. I enjoy learning continuously, collaborating with ambitious teams, and building products that create meaningful impact.

    When I need a break from the keyboard, you'll often find me immersed in the expansive realms of open-world fantasy RPGs, like the Elder Scrolls series. However, lately I've been thoroughly enjoying Black Myth Wukong whenever I'm free.

    Next ->
    Author Photo

  • Projects

    Mongoose-Footprints

    An NPM package, which is a plugin for the popular library 'mongoose'. It can log changes in MongoDB documents made using mongoose methods. If used on Mongoose models, any changes made to that model will be logged as a document of the _Footprint model. Footprints will contain an array that shows every change made to the document, along with the old and new document bodies. The plugin can be used to create an audit log for different applications. Jest has been used for thorough unit testing of the plugin.

    • Node.js
    • Mongoose
    • MongoDB
    • Jest (Unit Testing)
    • Mongoose-Footprints

    Tamashi

    • Anitrack
    • Anitrack

    A discord bot that can follow what a Spotify user is listening to and play it in a voice channel so everyone can listen in simultaneously. Although this feature already exists in Discord with the Spotify integration (Listen Along), only premium Spotify users get to use it. This bot can be a viable alternative. The bot will require you to authorize it to access your Spotify data using OAuth, after which it can track what you're playing by pinging the Spotify API in intervals.

    • Node.js
    • Redis
    • Discord.js
    • MongoDB
    • Docker
    • Express.js
    • Spotify API

    Genetic Algorithm for Path Planning

    An implementation of a modified genetic algorithm used in finding the shortest path from one point to another with some obstacles in between using the path points available throughout the space. I've used Matplotlib to show the simulation. The obstacles and path points are generated randomly on each run.

    The modification made improves the speed of chromosome generation by preventing the formation of lethal chromosomes. A matrix is used to cross out any paths that aren't valid and prevents their creation. Thus, every chromosome is a valid solution.

    • Python 3
    • Shapely
    • Matplotlib
    Genetic
    Next ->

  • Experience


    • Currently working as a Backend Engineer at Airwork AI, where I develop and maintain scalable REST APIs, search systems, and backend infrastructure using TypeScript, MongoDB, Express and Elasticsearch.
    • Worked as a Frontend Engineer at Polymath, building responsive Next.js interfaces and UI components for an AI chatbot platform serving thousands of users.
    • Worked as an intern at bKash in the Product & Technology department, contributing to internal SSO authentication system prototypes using Java and Spring Boot.
    • Worked as a student tutor of CS at BRAC University for Summer and Fall of 2021 to help over 60 students with their coursework
    • Secured the 1st runner up position at BRACU Intra University Programming Contest (Fall 2019) as a team of 3 to solve programming problems
    • Collaborated with 3 other students to conduct research in multimodal emotion recognition using different heterogeneous ensembles of machine learning models, achieving a classification accuracy of ~80% on 4 emotion classes. The research paper has been published in the 2022 25th International Conference on Computer and Information Technology (ICCIT).

    Next ->

  • Contact


    You can get in touch with me by: