Teaching Assistant

• Served as a teaching assistant for the Programming Language Paradigms course and the Database Design course.
• Held weekly office hours and graded assignments.

Software Engineer (AI/ML)

• Developed a knowledge mining solution utilizing Azure Cognitive Services and the OpenAI GPT model API.

Machine Learning Engineer

• Trained and developed novel Bangla word-level Optical Character Recognition (OCR) and Bangla Optical Handwriting Recognition (OHR) deep learning models as part of the Enhancement of Bangla Language in ICT through Research & Development (EBLICT) project, under the Ministry of ICT, Government of Bangladesh.
• Deployed multi-framework models for inference serving on both CPU and GPU servers using NVIDIA Triton Inference Server and Python web frameworks such as FastAPI and Flask.
• Conducted Document Denoising efforts to improve the accuracy of word segmentation and recognition processes.
• Contributed to the development of a Bangla Text-to-Speech deep learning model, enabling natural-sounding voice output from Bangla text.
• Customized a Natural Language Processing (NLP) project to identify similarities between doctor notes in natural language and SNOMED CT database concepts, leveraging state-of-the-art language models.

Research Assistant (RA)

• Explored deep learning techniques using synthetic data to address the scarcity of real-world data.
• Developed and studied quantized models to achieve a reduced memory footprint and faster inference on CPU.

Undergraduate Teaching Assistant (UGTA)

• Served as a teaching assistant for the Object-Oriented Programming course and the Microprocessor Interfacing & Embedded Systems course.
• Held weekly office hours and graded assignments.

The University of Texas at Dallas

CGPA: 3.83 on a scale of 4.00

North South University

CGPA: 3.74 on a scale of 4.00 (Graduated magna cum laude.)

A Multifaceted Evaluation of Representation of Graphemes for Practically Effective Bangla OCR

Abstract: Bangla Optical Character Recognition (OCR) poses a unique challenge due to the presence of hundreds of diverse conjunct characters formed by the combination of two or more letters. In this paper, we propose two novel grapheme representation methods that improve the recognition of these conjunct characters and the overall performance of OCR in Bangla. We have utilized the popular Convolutional Recurrent Neural Network architecture and implemented our grapheme representation strategies to design the final labels of the model. Due to the absence of a large-scale Bangla word-level printed dataset, we created a synthetically generated Bangla corpus containing 2 million samples that are representative and sufficiently varied in terms of fonts, domain, and vocabulary size to train our Bangla OCR model. To test the various aspects of our model, we have also created 6 test protocols. Finally, to establish the generalizability of our grapheme representation methods, we have performed training and testing on external handwriting datasets. Experimental results proved the effectiveness of our novel approach. Furthermore, our synthetically generated training dataset and the test protocols are made available to serve as benchmarks for future Bangla OCR research.

Rethinking Task-Incremental Learning Baselines

Abstract: It is common to have continuous streams of new data that need to be introduced in the system in real-world applications. The model needs to learn newly added capabilities (future tasks) while retaining the old knowledge (past tasks). Incremental learning has recently become increasingly appealing for this problem. Task-incremental learning is a kind of incremental learning where task identity of newly included task (a set of classes) remains known during inference. A common goal of task-incremental methods is to design a network that can operate on minimal size, maintaining decent performance. To manage the stability-plasticity dilemma, different methods utilize replay memory of past tasks, specialized hardware, regularization monitoring etc. However, these methods are still less memory efficient in terms of architecture growth or input data costs. In this study, we present a simple yet effective adjustment network (SAN) for task incremental learning that achieves near state-of-the-art performance while using minimal architectural size without using memory instances compared to previous state-of-the-art approaches. We investigate this approach on both 3D point cloud object (ModelNet40) and 2D image (CIFAR10, CIFAR100, MiniImageNet, MNIST, PermutedMNIST, notMNIST, SVHN, and FashionMNIST) recognition tasks and establish a strong baseline result for a fair comparison with existing methods. On both 2D and 3D domains, we also observe that SAN is primarily unaffected by different task orders in a task-incremental setting.

Analysis of Bangla Keyboard Layouts Based on Keystroke Dynamics

Abstract: Keyboards are the primary devices for interaction with computer platforms and have been a central topic in HCI research for decades. The study of their layout design is important in deciding their efficiency, practicality and adoption. Multiple keyboard layouts have been developed for Bangla without any rigorous study for their comparison. In this paper, we take a quantitative data-driven approach to compare their efficiency. Our evaluation strategy is based on the key-pair stroke timing with data collected from standard QWERTY English keyboards. Our experiments conclude that the Bijoy keyboard layout is the most efficient design for Bangla among the four layouts studied. This quantitative approach can lay the groundwork for further study of these layouts based on other criteria.

Mathematics for Machine Learning and Data Science Specialization

Courses covered Linear Algebra for Machine Learning and Data Science, Calculus for Machine Learning and Data Science and Probability & Statistics for Machine Learning & Data Science.

Machine Learning Specialization

Courses covered Linear Regression, Logistic Regression, Regularization, Gradient Descent, Artificial Neural Network, Decision Trees, Tree Ensembles, Xgboost, Recommender Systems, K-means Algorithm and Reinforcement Learning.

Deep Learning Specialization

Courses covered Deep Learning, Optimization Techniques, Convolutional Neural Networks and Sequence Models.

Automate the Boring Stuff with Python Programming

Learned basic task automation with python, word, pdf and excel document parsing, desktop automation, web scrapping and email automation, regular expression etc.

Speaker Verification

Speaker verification system using deep learning model that can verify the identity of a speaker from raw waveform.

Sketch Colorization

Coloring black and white sketches using generative adversarial network.

Ashbab e-commerce app

An e-commerce android application that uses augmented reality to preview a product before buying.

Line Follower Robot with Cave Navigation

An autonomous line-follower robot that follows a line using an infrared sensor. Also capable of navigating through a cave in the absence of a line.

Vice Chair

Planned chapter activities and conducted weekly meetings. Took basic robotics workshops as an instructor.

Content Writing Coordinator

Planned events, wrote event proposals and event reports. Participated in all the workshops and seminars organized by the student branch.