S
Portrait of Sajidur Rahman Sajid

AI Engineer & Full-Stack Developer

Sajidur Rahman Sajid

I build full-stack systems and production-ready AI/ML solutions, turning complex problems into reliable software for the real world.

3.91CGPA / 4.00
5xDean's Award
15+Projects Built
2+Years Building
  • TypeScript
  • Python
  • JavaScript
  • React
  • Next.js
  • Node.js
  • FastAPI
  • NestJS
  • Express
  • PostgreSQL
  • MongoDB
  • Redis
  • SQLAlchemy
  • Docker
  • PyTorch
  • TensorFlow
  • Scikit-learn
  • Hugging Face
  • Pandas
  • NumPy
  • OpenCV
  • Tailwind CSS
  • Git
  • Linux
  • Vercel

[ 01 ] · About

The world is part of the work.

  • CURRENTLY BUILDING·SOFTWARE SYSTEMS
  • WORKING ACROSS·FULL-STACK + AI/ML
  • BASED IN DHAKA, BANGLADESH·UTC+6

I'm Sajid, a software engineer in Dhaka. I work across full-stack systems and AI/ML, building software that turns complicated problems into something practical and reliable.

The work reaches past my desk. There's a growing footprint across America and Europe, and I like software that travels well beyond where it was built.

3 regions tracked · footprint growing

loading globe…
drag to explore

Bangladesh

Home / base

[ 02 ] · Experience

Engineering scalable applications and enterprise production software.

May 2026 - PresentPresent

Ledgercross

Software Developer Trainee

Engineering production software across the stack for enterprise finance, working with APIs, asynchronous processing, data layers, automated testing, cloud infrastructure, and system reliability.

Feb 2026 - Apr 2026Complete

Bangladesh Software Solution

Software Engineering Intern

Developed responsive frontend interfaces with HTML5, CSS3, and modern JavaScript (ES6+), integrated them with backend RESTful APIs with robust error handling, and worked in agile, Git-based workflows.

[ 03 ] · Projects

Built things. Broke things. Shipped things.

View all projects

active build · professional

01

AI/ML Engineer · AI extraction service

AI-powered invoice processing platform that extracts, validates, and analyzes invoices while detecting duplicates and anomalies to streamline financial document workflows.

FastAPIPythonMindee OCRPostgreSQLRedis StreamsCloudflare R2

active build · professional

02

Full-stack Engineer

Production inventory and point-of-sale system for a retail operation: a phone barcode scanner feeds a web app that manages products, stock, sales, and revenue analytics over the shop's local network.

KotlinJetpack ComposeNext.jsNeon PostgreSQLPrismaCloudflare R2
Live DemoCode

active build · professional

03

Full-Stack / AI Engineer

Privacy-first legal research workspace that ingests case documents, ranks search results by relevance, with an AI summary and citation layer designed for Phase 2.

FastAPISQLAlchemySQLiteNext.js
Live DemoCode

Archive

More from the build log.

COMPLETE

Turf booking platform that lets players discover and reserve grounds on a Google Map while owners publish and manage slots with overlapping-time checks.

ReactExpressMongoDB AtlasRedux ToolkitGoogle Maps
GitHubLive Demo
Neuro-Screen: Hybrid CatBoost + ANN framework that screens cognitive-impairment risk in insomniac university students from a self-reported lifestyle questionnaire, with explainable predictions.
Research

Neuro-Screen

COMPLETE

Hybrid CatBoost + ANN framework that screens cognitive-impairment risk in insomniac university students from a self-reported lifestyle questionnaire, with explainable predictions.

PythonCatBoostPyTorchStreamlit
FinBERT: Comparative study proving domain-tuned transformers outperform generic models on financial sentiment, with a full fine-tuning-to-evaluation pipeline in a single notebook.
NLP

FinBERT

COMPLETE

Comparative study proving domain-tuned transformers outperform generic models on financial sentiment, with a full fine-tuning-to-evaluation pipeline in a single notebook.

BERTTransformersPyTorchNLP
CodingVibes: Java Swing course-storefront app with admin and user login, course browsing and a cart, storing accounts in flat files.
Desktop

CodingVibes

COMPLETE

Java Swing course-storefront app with admin and user login, course browsing and a cart, storing accounts in flat files.

JavaSwingFile I/O
Face Recognition System: Real-time face identification pipeline that detects faces with a Haar cascade, trains an OpenCV LBPH recognizer on a known set, and identifies people live, from dataset to inference in one reproducible notebook.
CVPR

Real-time face identification pipeline that detects faces with a Haar cascade, trains an OpenCV LBPH recognizer on a known set, and identifies people live, from dataset to inference in one reproducible notebook.

PythonOpenCVLBPH
City Simulator: Animated OpenGL city scene built by a team of three; I led the animation work: a day/night toggle, particle rain and snow, a traffic-light state machine, and moving vehicles and pedestrians.
Graphics

Animated OpenGL city scene built by a team of three; I led the animation work: a day/night toggle, particle rain and snow, a traffic-light state machine, and moving vehicles and pedestrians.

C++OpenGLGLUT
Spark Powerhouse: Windows gym management application separating member and admin workflows for memberships, payments, and daily records in a role-aware system.
Desktop

Windows gym management application separating member and admin workflows for memberships, payments, and daily records in a role-aware system.

C#WinFormsSQL Server
Employee & Family Registry: Employee registry with spouse and children records, case-insensitive search across name, NID and department, and on-demand PDF CV and list exports from one ASP.NET Core API.
Desktop

Employee registry with spouse and children records, case-insensitive search across name, NID and department, and on-demand PDF CV and list exports from one ASP.NET Core API.

ASP.NET CoreEF CorePostgreSQLReact
Bookstore Database: Relational design for an online bookstore, normalized through 3NF, with an ER diagram and catalog, order and payment queries.
Database

Relational design for an online bookstore, normalized through 3NF, with an ER diagram and catalog, order and payment queries.

SQLNormalizationER Diagram

[ 04 ] · Skills

My current engineering toolkit.

  • TypeScript
  • JavaScript
  • Python
  • Java
  • C++
  • C#
  • .NET
  • React
  • Next.js
  • Node.js
  • Express
  • NestJS
  • Tailwind CSS
  • HTML5
  • CSS3
  • FastAPI
  • PostgreSQL
  • MongoDB
  • Redis
  • PyTorch
  • TensorFlow
  • Scikit-learn
  • Hugging Face
  • OpenCV
  • Docker
  • Git
  • Linux
  • Vercel
  • Cloudflare
  • Solidity

Backend

PythonFastAPINode.jsExpressNestJSPostgreSQLSQLAlchemyRedis StreamsDockerREST

AI / ML

PyTorchCatBoostScikit-learnHugging FacePandasNumPyTransformersOpenCVSHAP / LIME

Frontend

ReactNext.jsTypeScriptTailwind CSSViteReduxFramer Motion

Data & Databases

MongoDBSQLiteSupabaseCloudflare R2

Blockchain

SolidityEthereumSmart ContractsHardhat

Tooling & Practice

GitpytestVercelLinuxNpm Workspaces

[ 05 ] · Education

The foundations underneath the stack.

AIUB campus at dusk: the dome building and the main academic buildingHighest qualification

American International University-Bangladesh

Bachelor of Science in Computer Science and Engineering

3.91
CGPA / 4.00
Dhaka, BangladeshSep 2022 - Apr 2026
5× Dean's Award
AIUB Merit Scholar
Duke of Edinburgh Bronze
Kurmitola High School & College logo

Kurmitola High School & College

Secondary School Certificate (SSC) · Science

Jan 2017 - Feb 2019

5.00
GPA / 5.00
Dhaka
BAF Shaheen College Dhaka logo

BAF Shaheen College Dhaka

Higher Secondary Certificate (HSC) · Science

Jun 2019 - Dec 2021

5.00
GPA / 5.00
Dhaka

[ 06 ] · Research

Where the models meet the messy real world.

Featured thesis

Live App

Neuro-Screen: A Hybrid Ensemble Framework for Detection of Cognitive Impairment in Insomniac University Students

Thesis that fuses a gradient-boosting classifier with a three-layer neural network by averaging their probability outputs. Trained on 2,237 survey responses from Bangladeshi university students, the ensemble beats every standalone model across all metrics and pinpoints the lifestyle factors that most strongly predict cognitive decline.

95.20%
accuracy
94.40%
precision
96.10%
recall
95.24%
F1
0.982
ROC-AUC
2,237
survey responses

[ 07 ] · cortex

Peek inside the machine.

system online
cortex loading_

[ 08 ] · Contact

Let's turn the idea into something real.

Open to opportunities

Location

Dhaka, Bangladesh

UTC+6 · remote-friendly

Availability

I'm open to software engineering roles across full-stack, platform, and applied ML, plus select freelance and product work if it's interesting.