MD Nahiduzzaman Dipu Python · Django · Machine Learning

Open to junior developer opportunities

MD Nahiduzzaman Dipu Building practical software with Python, Django & AI.

I'm MD Nahiduzzaman Dipu, a Computer Science graduate from Bangladesh focused on backend development, machine learning, and building practical software solutions.

POTHPRADARSHAK · ILLUSTRATION FRAME 0001
সামনে একটু বাঁ দিকে একজন মানুষ। “Person ahead, slightly left.”

A drawing of what my thesis system does — not a live model. See the real results →

Based in
Bangladesh UTC+6
Education
B.Sc. in Computer Science and Engineering
Experience
Python/Django Intern, O'dell Tech Ltd.
Looking for
Junior Python · Django · ML roles

01 About me

I like software that does a job for someone.

I studied Computer Science and Engineering at Green University of Bangladesh. What kept me interested through the degree wasn't any single subject — it was the moment a piece of software stops being an assignment and starts being something a person can actually use.

Python is where I'm most at home. As a Python/Django developer intern at O'dell Tech Ltd. I worked on a point-of-sale system built to a written business specification: role-based access, sales that run inside database transactions, and a test suite of 281 tests. That project taught me how much of backend work is about being correct every time — with money, with stock, with permissions.

Alongside the backend work I trained in machine learning and deep learning at Creative IT Institute, and my thesis, PothPradarshak, put that into practice: I built and verified a dataset, trained a YOLO object detector, measured it honestly class by class, and wrapped it in a real-time application that gives spoken guidance in Bengali and English. CitizenCare, a public-safety platform with a YOLOv8 camera service, sits between the two worlds.

I'm looking for my first full-time role as a junior Python or Django developer, or a junior machine learning engineer — somewhere I can ship real features, learn from experienced engineers, and keep getting better at both.

Outside of code

  • Photography. Portrait and street photography. Framing a shot is a lot like framing a problem.
  • Design. Visual identity and layout work in Photoshop and Illustrator, from before I moved into software.
  • Travelling. Seeing places outside the city — and photographing them.

Languages

  • BengaliNative
  • EnglishProfessional working proficiency
  • HindiConversational
  • UrduConversational
  • GermanBeginner (A1)

02 Technical skills

Skills, with the evidence attached.

No percentages or progress bars. Each skill links to the projects on this site that use it — choose a project to see what it took.

Show skills used in

Backend

Python first — Django for complete applications, FastAPI and Flask for model services.

Machine Learning

Training, evaluating and serving models — computer vision most of all.

2 projects on this site that use it — select a skill to see which from training or coursework — no public project yet

03 Featured projects

Real projects, with their results and their limits.

Screenshots are from the projects' own repositories. Each project has a full write-up: the problem, the architecture, how it was built, what was measured and what still doesn't work.

Academic thesis2026

PothPradarshak

A YOLO-based vision assistant that turns Dhaka footpath scenes into spoken Bengali guidance

Problem. Off-the-shelf detectors are trained on datasets that barely represent a Dhaka footpath — parked vehicles on the pavement, broken curbs, street obstacles — and their output is a list of boxes, not something a person can act on …

  • YOLO11n detector trained on 2,782 hand-verified Dhaka footpath images
  • Turns each detection into left / centre / right and near / far, then ranks by risk
  • Spoken guidance in Bengali and English, with cooldowns so it never chatters
  • Live camera, image upload and video modes served through FastAPI

My roleThesis author — dataset, training, evaluation and the real-time application

  • Python
  • PyTorch
  • Ultralytics YOLO11
  • YOLO-World
  • FastAPI
  • edge-tts
  • pyttsx3
  • CUDA

No public live demo — runs locally from the repository

Project2026

CitizenCare

A public-safety platform: citizen crime reports, real-time SOS, and YOLOv8 CCTV analysis

Problem. Incident reporting is usually slow, unverified and disconnected from the people who respond. The platform needed three very different users — citizens, police and administrators — to share one system without any of them being able to see …

  • Citizens report incidents with photos, video and a map pin, and can send an SOS
  • Police verify reports, answer SOS alerts and watch camera feeds from one command view
  • A Flask + YOLOv8 service analyses about one CCTV frame per second
  • JWT auth, bcrypt, rate limits, an audit log and role-checked Socket.IO channels

My roleDeveloper — published and documented in my GitHub repository

  • React
  • Vite
  • Tailwind CSS
  • Node.js
  • Express
  • MongoDB
  • Socket.IO
  • JWT

No public live demo — runs locally from the repository

Internship project2026

POS System — O'dell Tech

A Django point-of-sale system from an internship at O'dell Tech, built to a written business specification

Problem. A shop till has to be right about money and stock every time, under concurrent use, and it has to stop a cashier from doing a manager's job. The specification set out 36 business requirements and twelve business …

  • Till workflow: product search, barcode scan, cart, discounts, tax and invoice
  • Sales run in atomic transactions with row locks, so two tills can't sell the last unit
  • Four roles mapped to named capabilities, checked on every view
  • 281 automated tests passing on both SQLite and MySQL

My rolePython/Django Developer Intern at O'dell Tech Ltd. (16 Aug – 19 Sep 2026)

  • Python
  • Django
  • MySQL
  • SQLite
  • JavaScript
  • HTML
  • CSS
  • WhiteNoise

No public live demo — the company deployment needs a staff login

More work

  1. 01 Meal Management System Shared-housing meal and expense tracking, built end to end ReactViteTailwind CSSNode.jsExpress Personal project 2026 Built
  2. 02 Jersportss A storefront for sportswear, designed and built end to end HTMLCSSJavaScriptGitHub Pages Personal project 2025 Archived
  3. 03 Design & Marketing Visual identity and campaign work from before the move into ML PhotoshopIllustratorSEOMeta Ads Training work 2023 Archived

04 Thesis & research

Teaching a camera to describe a footpath.

Undergraduate thesis · 2026 Green University of Bangladesh Research prototype

PothPradarshak: A Real-Time Vision-Based Framework for Object Detection and Navigation Assistance for Visually Impaired Individuals

Abstract

PothPradarshak (পথপ্রদর্শক, “the one who shows the way”) is my undergraduate thesis prototype: a real-time system that watches a camera feed, detects people, vehicles and footpath hazards, works out where they are relative to the walker, and speaks short guidance in Bengali or English. It is a research prototype, not a certified mobility aid.

Where it stands
The full pipeline works end to end: live camera, uploaded image and uploaded video all run through detection, spatial reasoning and Bengali/English voice guidance at about 25 FPS on a laptop RTX 3050 Ti. On the held-out test split the trained model reaches precision 0.645, recall 0.404, mAP@50 0.419 and mAP@50-95 0.297 overall. It is genuinely good at the two well-represented classes — person (mAP@50 0.837) and parked vehicle (0.781) — and weaker on the rare classes, which have few examples.
mAP@50 by class Results for seven classes on the held-out test split — YOLO11n, 100 epochs, 640 px.
  • person 0.837 n = 878
  • parked_vehicle 0.781 n = 1360
  • curb 0.359 n = 20
  • street_obstacle 0.295 n = 99
  • pole 0.268 n = 9
  • tree 0.237 n = 65
  • stairs 0.995 n = 2

Hatched bars have fewer than 10 test boxes. Stairs (two test boxes) and pole (nine) are too small a sample to read as evidence — a high score there is noise, not a result.

  1. 01CameraThreaded capture, newest frame wins
  2. 02DetectionYOLO11n, footpath classes
  3. 03Spatial analysisLeft / centre / right · near / medium / far
  4. 04Navigation enginePriority score, events, cooldown
  5. 05VoiceBengali (edge-tts) · English (pyttsx3)
  6. 06Web UIFastAPI · MJPEG stream · JSON API

Currently learning

Self-study

Django REST Framework

Moving from server-rendered views to clean, documented JSON APIs.

Research

Closing the class-imbalance gap

Collecting more verified examples of PothPradarshak's rare classes, such as stairs and curbs.

Self-study

Model deployment

Serving models outside a laptop GPU: latency budgets, containers and embedded targets.

05 Experience

One internship, and the path that led to it.

I'm early in my career: one internship so far, no full-time role yet. Everything below is labelled for what it was.

Internship

Python/Django Developer Intern

O'dell Tech Ltd.

Worked on the company's point-of-sale system for O'dell Tech Shopping in Python and Django, built to a written Business Requirement Specification.

  • Worked in a Django codebase organised around a services layer, where all money and stock logic lives — never in views.
  • Sales run inside atomic database transactions with row locks; every stock change is written to a movement ledger.
  • Role-based access for four roles (Administrator, Manager, Cashier, Inventory Staff) through named capabilities checked on every view.
  • Bulk product import from CSV / Excel with a preview step before anything is written.
  • An automated suite of 281 tests, run against both SQLite and MySQL.
  • Python
  • Django
  • MySQL
  • SQLite
  • JavaScript
  • HTML
  • CSS

The path so far

  1. 2022
    Education

    Started a B.Sc. in CSE

    Began Computer Science and Engineering at Green University of Bangladesh, after a science background at HSC level.

  2. 2023
    Training

    Design and digital marketing

    Trained at SR Dream IT in graphics design and digital marketing — it taught me to think about the person on the other side of the screen.

  3. 2023 – 2025
    Personal projects

    Building for the web

    Moved into web development, including Jersportss — a storefront designed, built and launched end to end.

  4. 2025 – 2026
    Training

    Machine Learning and Deep Learning training

    Trained at Creative IT Institute and started building models with real evaluation discipline rather than tutorials.

  5. 2026
    Academic research

    Computer vision thesis

    PothPradarshak: built the dataset, trained and evaluated a YOLO11 detector, and shipped a real-time app that speaks Bengali navigation guidance.

  6. 2026
    Project

    CitizenCare

    A public-safety platform with citizen reporting, real-time SOS and a YOLOv8 CCTV analysis service.

  7. 2026
    Education

    Graduated in CSE

    Completed the B.Sc. in Computer Science and Engineering at Green University of Bangladesh.

  8. Aug – Sep 2026
    Internship

    Python/Django Developer Intern, O'dell Tech Ltd.

    Worked on the company's Django point-of-sale system: role-based access, atomic sales with row locking, and a 281-test suite.

06 Education & training

Education & training.

Education

  • 2022 – 2026

    B.Sc. in Computer Science and Engineering

    Green University of Bangladesh

    Completed. Coursework and independent work in software engineering, machine learning and computer vision, with a thesis in assistive computer vision (PothPradarshak).

  • 2019 – 2021

    Higher Secondary Certificate

    Maijdee City College, Noakhali

    Science group. First exposure to programming and to the web fundamentals that led into a CSE degree.

  • 2018 – 2019

    Secondary School Certificate

    Harinarayanpur Union High School, Noakhali

    Science group.

Training

  • 2025 – 2026

    Machine Learning and Deep Learning

    Creative IT Institute

    Training in machine learning and deep learning with Python: data preparation, feature engineering, model training and evaluation, and neural networks.

  • 2023

    Digital Marketing & Graphics Design

    SR Dream IT

    Hands-on training in SEO, social media marketing, and visual content production with Adobe Photoshop and Illustrator.

07 Career opportunities

What I'm looking for.

I'm actively applying for junior roles. If your team is hiring for one of these, I'd like to hear from you.

Where
Bangladesh — on-site, hybrid or remote
Level
Junior / entry-level, full-time
Available
Now

08 Contact

Let's talk.

I'm looking for junior Python, Django or machine learning roles, and entry-level AI work. Based in Bangladesh — open to on-site, hybrid and remote positions. The form sends straight to my inbox; I reply from mdnahiduzzamandipu@gmail.com.

0 / 4000 · at least 20 characters

Required. Your details are only used to reply to you. Prefer email? mdnahiduzzamandipu@gmail.com

↑↓ to move · Enter to run · Esc to close