Help & Supportcareers@gravityer.com
ML Engineer with 1+ years in Data Science & Web Development
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Highly motivated and results-driven IT Executive and Student Tutor with 1.6 years of combined experience in system administration, IT asset management, and academic mentorship. Proficient in Python, Machine Learning, Deep Learning, and Web Development, with a strong foundation in data science and problem-solving. Proven ability to manage complex systems, streamline operations, and guide students in technical subjects, consistently improving outcomes through analytical and collaborative approaches.
BRAC University
Bachelor of Science · Computer Science
August 1, 2021 – June 30, 2025
Hamdard Public College
Higher Secondary School Certificate
June 1, 2018 – May 31, 2020
United Aygaz LPG LTD.
IT Executive
August 1, 2025 – April 1, 2026
Dhaka, Dhaka Division, Bangladesh
BRAC University
Student Tutor / Teacher Assistant
June 1, 2024 – May 1, 2025
Dhaka, Dhaka Division, Bangladesh
Sequence Learning with Recurrent Neural Networks
March 1, 2025 – May 1, 2025
Developed and compared RNN, LSTM, GRU, and BiLSTM models for sequence-based POS tagging; GRU gained 88.26% accuracy, outperforming LSTM by 5.7%. Optimized GRU using systematic hyperparameter tuning and dropout regularization, increasing validation F1-score from 82.4% to 86.9%. Visualized model performance using confusion matrices and F1-scores, achieving a macro F1-score of 85.3% across 10 POS classes.
View ProjectPlant Disease Segmentation Using U-Net
October 1, 2024 – November 1, 2024
Create a U-Net model achieving 91.3% accuracy and a MeanIoU of 88.6% on a dataset of 3,000+ annotated leaf images. Applied advanced data augmentation techniques, boosting validation accuracy by 9.4%. Designed a custom focal loss function that improved segmentation of small diseased regions by 15.2%.
View ProjectAnomaly Detection From CCTV Footage (Thesis)
July 1, 2024 – December 1, 2024
Implemented TimeSformer (67%), Uniformer (57%), VideoMAE (48%) for anomaly detection. Engineer a smart surveillance system that Attained a 62.5% average precision and reduced false positives by 14% using frame-level predictions. Extracted frame-wise embeddings from 12,000+ video segments, enabling detection of abnormal activity with an F1-score of 64.2%.
View ProjectVideo Description Generator Using HumanLava (Thesis)
July 1, 2024 – December 1, 2024
Design a video captioning system using the HumanLava model, achieving 84.2% accuracy and reducing captioning errors by 21% compared to baseline models. Engineered a pipeline combining NLP and ML, improving BLEU-4 scores by 18.5% over standard LSTM baselines. Processed over 7,500 labeled video clips using an encoder-decoder RNN architecture, generating human-like captions with a METEOR score of 76.4%.
View ProjectJerry Cheese Catching Game
February 1, 2024 – March 1, 2024
Construct a 2D game using OpenGL with collision detection and real-time input handling. Optimized for smooth rendering and interaction using graphical computations. Implemented score-tracking and level progression for better player engagement.
View ProjectMachine Learning for Brain Stroke Prediction
October 1, 2023 – December 1, 2024
Deploy a predictive pipeline using Logistic Regression, Random Forest, and Gradient Boosting; Accomplished 92.1% ROC-AUC and 89.7% F1-score on test data. Used GridSearchCV to optimize hyperparameters, improving model accuracy by 7.4%. Applied SMOTE to balance classes with a 3:1 skew, enhancing recall by 12%.
View ProjectSmart Educational Platform
April 1, 2023 – October 1, 2023
Built SmartEduHub to manage academic operations with secure login and role-based access. Integrated payment system, communication tools, and class/exam routines. Implemented dashboards for Admin, Teacher, and Student with personalized content and access.
View ProjectCompleted Courses in Unsupervised Learning Reinforcement Learning
Coursera
June 1, 2026 – Present
Certified for Completing Speak Bold Workshop
Unknown
June 1, 2026 – Present
Completed Aspire Leadership Program 2025
Unknown
January 1, 2025 – Present
Cultural Fit Analysis
The candidate's academic projects show a strong interest in diverse applications of ML, including healthcare, surveillance, and agriculture, indicating a broad intellectual curiosity. Participation in extracurricular activities like the Robotics Club and various competitions suggests a proactive, collaborative, and competitive spirit. The IT Executive role, though brief, indicates an ability to work in a corporate environment. The candidate's involvement in tutoring and leadership programs suggests a willingness to contribute to a team and mentor others, which aligns well with a collaborative work culture.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through project work and academic achievements. Their experience as a student tutor highlights communication and mentorship abilities. Participation in leadership programs and business case competitions suggests teamwork, leadership potential, and adaptability. The IT Executive role, while not directly technical in ML, shows operational experience in system administration and asset management, which can be beneficial for understanding deployment and infrastructure needs.