CSE Graduate  ·  AI Automation Enthusiast  ·  Healthcare AI Researcher  ·  Python Developer
Building practical AI systems for automation, healthcare, document understanding, and intelligent workflows

Undergraduate thesis on healthcare document understanding for low-resource Bangla-English handwritten prescriptions.

This thesis developed an end-to-end AI pipeline for recognizing handwritten medicine names from bilingual prescriptions and transforming extracted information into patient-friendly summaries.

Key Contributions

• Converted prescription-region annotations into OCR-ready medicine-name crops.
• Fine-tuned a Transformer-based OCR model, TrOCR, for handwritten medicine-name recognition.
• Applied lexicon-guided Levenshtein post-processing to reduce OCR errors.
• Integrated corrected medicine names with structured medicine knowledge sources.
• Generated deterministic Bangla-English patient-centric summaries with safety disclaimers.
• Evaluated performance using CER, WER, exact-match accuracy, and error analysis.

Key Areas: Healthcare AI OCR TrOCR Computer Vision NLP Medical Document Understanding Low-Resource AI Patient-Centered AI Status: Completed Undergraduate Thesis

Course-related research on biomedical signal classification for heartbeat/phonocardiogram analysis.

This project compared classical machine learning, standalone deep learning, CNN-RNN hybrid models, and ensemble approaches for heartbeat sound classification.

Key Contributions

• Preprocessed heartbeat sound recordings using resampling, silence trimming, fixed-window segmentation, and log-mel spectrogram generation.
• Extracted handcrafted audio features for classical ML models.
• Benchmarked models including SVM, KNN, Random Forest, GBM, XGBoost, LightGBM, Decision Tree, Naïve Bayes, CNN, LSTM, Bi-LSTM, and GRU.
• Designed and evaluated CNN+LSTM, CNN+Bi-LSTM, and CNN+GRU hybrid architectures.
• Evaluated performance using accuracy, precision, recall, F1-score, confusion matrix, and training/validation curves.

Key Areas: Biomedical Signal Processing Heartbeat Sound Classification Machine Learning Deep Learning CNN LSTM GRU Hybrid Models Healthcare AI Status: Completed Course-Related Research

Team-based coursework project on supervised classification using structured facial-feature data.

This project developed a simple ANN-based binary classification system using a structured CSV dataset containing facial-feature variables.

Key Contributions

• Worked with a dataset of 5,001 samples and 7 facial-feature variables.
• Applied preprocessing, feature shaping, train-test splitting, and supervised model training.
• Built an ANN with ReLU hidden layers, Sigmoid output, Binary Crossentropy loss, Adam optimizer, and early stopping.
• Evaluated model performance using test accuracy, confusion matrix, precision, recall, F1-score, and ROC curve.

Key Areas: Artificial Neural Networks Pattern Recognition Supervised Learning Binary Classification TensorFlow Keras Status: Completed Team-Based Coursework Project

During my undergraduate study, I completed several academic projects that strengthened my foundation in software development, object-oriented programming, embedded systems, data analysis, and database-driven application design.

• ATM/CRM Machine System Management Developed a machine-management system using Object-Oriented Programming concepts, focusing on modular design, class-based structure, transaction handling, and system logic.

• Automated Railway Gate Control System Built an embedded-system prototype using Arduino UNO and ultrasonic sensors to automate railway gate operation based on object/train detection.

• Local Transport System Mismanagement Analysis Conducted a statistical analysis of public perceptions and proposals on local transport mismanagement using data from 250 participants, including drivers, passengers, and local residents.

• Timely Care — Doctor Appointment Management Software Developed a doctor appointment management system to support patient scheduling, appointment tracking, and healthcare-service organization.

Key Areas: OOP Embedded Systems Arduino Sensor-Based Automation Data Analysis Software Engineering Database Systems Healthcare Software

AI, ML & Research

AI Automation & LLM Tools

Web & Backend

Data, Databases & Tools

Security & Responsible AI Interests

AI tool that scores a resume against a job description and suggests improvements.

This project allows users to upload a PDF resume and paste a job description. The system analyzes the resume, calculates a match score, identifies missing keywords, and provides actionable suggestions to improve the resume for a target role.

Tech: Python OpenAI API PyPDF2 Streamlit Status: Completed

• PCIU Merit Scholarship Award for Semester GPA 4.00 — Fall 2022, Summer 2023, Fall 2024, Fall 2025
• Participated in ICPC Regional Preliminary Contest — 2022, 2023
• Class Representative at PCIU Computer Club — 2024
• Member, Bangladesh National Cadet Corps — 2017–2018

✅ AI chatbot for websites or business workflows ✅ Resume analysis and document intelligence tools ✅ RAG-based question-answering systems over documents ✅ n8n / Make.com workflow automation ✅ Data scraping and AI analysis pipelines ✅ Django or FastAPI backend with AI features ✅ Healthcare/document-processing AI prototypes

I'm actively looking for opportunities in AI engineering, automation development, IT, software development, and research assistant roles across Bangladesh and remote positions worldwide.

I am also interested in graduate research opportunities related to human-centered healthcare AI, medical document understanding, biomedical signal analysis, trustworthy AI, low-resource clinical AI, and robust model adaptation.

• 💼 LinkedIn: linkedin.com/in/rakib-mohammad-ammar-25a966390
• 📧 Email: rakibmdammar@gmail.com
• 📱 WhatsApp / Phone: +8801867598104 (for professional enquiries)

"I don't just learn AI tools — I build practical systems that solve real problems."