I'm Maulik Gajera, a B.Tech Artificial Intelligence student at Gandhinagar University, focused on machine learning, deep learning, and applied data science.
My work spans the full modeling lifecycle: exploratory data analysis, preprocessing, feature engineering, model selection, evaluation, and packaging models behind an API. I enjoy working across problem types — computer vision, NLP, and classical tabular ML — and I publish datasets and notebooks on Kaggle to share that work publicly.
| Core Area | Focus |
|---|---|
| Machine Learning | Regression, classification, time-series forecasting, feature engineering, model evaluation |
| Deep Learning | CNNs for image classification, LSTM & Transformer models (DistilBERT, MiniLM) for NLP |
| Data Science | EDA, data cleaning, visualization, Power BI / Tableau reporting |
| Deployment | FastAPI, Flask, Docker |
| Kaggle | Dataset publishing, applied ML notebooks |
| Domain | Details |
|---|---|
| Machine Learning | Regression, classification, ensemble methods (Random Forest, SVM, Decision Tree), model comparison |
| Deep Learning | CNN-based image classification; LSTM and Transformer (DistilBERT, MiniLM) architectures for text |
| NLP | Text preprocessing, tokenization, sentiment classification, evaluation via F1 / ROC-AUC |
| Data Science | EDA, feature engineering, data visualization, dashboarding |
| MLOps Basics | Model packaging with FastAPI / Flask, containerization with Docker |
Review Sentiment Analysis
An end-to-end NLP pipeline for classifying review sentiment, comparing a classical sequence model against transformer-based encoders.
| Attribute | Details |
|---|---|
| Stack | Python, LSTM, DistilBERT, MiniLM, Scikit-learn |
| Pipeline | Text preprocessing, tokenization, model training, evaluation |
| Performance | 92% accuracy, evaluated with precision, recall, F1-score, and ROC-AUC |
| Repository | View on GitHub |
Student Placement Prediction
A classification system comparing multiple algorithms to predict student placement outcomes from academic and demographic features.
| Attribute | Details |
|---|---|
| Stack | Python, Logistic Regression, Decision Tree, Random Forest, SVM |
| Pipeline | EDA, feature engineering, multi-model comparison |
| Performance | Best result with Random Forest after comparative evaluation |
| Repository | View on GitHub |
Pokémon Image Classification
A CNN-based image classifier distinguishing Pokémon species, with an emphasis on preprocessing and augmentation to improve generalization.
| Attribute | Details |
|---|---|
| Stack | Python, CNN, image augmentation |
| Pipeline | Preprocessing, augmentation, hyperparameter tuning |
| Performance | Evaluated with confusion matrices, accuracy, and loss curves |
| Repository | View on GitHub |
Dog Emotion Classification
A computer vision project classifying dog emotions from images using transfer learning, comparing two pretrained CNN backbones under a shared classification head.
| Attribute | Details |
|---|---|
| Stack | Python, PyTorch, MobileNetV2, EfficientNetB0 |
| Pipeline | Data loading & preprocessing → transfer learning → 2-phase training → evaluation → model comparison |
| Evaluation | Accuracy, precision, recall, F1-score, ROC-AUC, confusion matrix |
| Repository | View on GitHub |
Smart Housing: AI-Powered Price Prediction (Kaggle)
A regression notebook estimating residential sale prices on the Ames, Iowa housing data (the "House Prices – Advanced Regression Techniques" dataset).
| Attribute | Details |
|---|---|
| Stack | Python, regression modeling |
| Dataset | House Prices – Advanced Regression Techniques (Ames Housing, 79 explanatory variables) |
| Notebook | View on Kaggle |
Gender Wage Gap Analysis (Kaggle)
An exploratory data analysis notebook examining pay disparity patterns using the "Mind the Gender Wage Gap" dataset.
| Attribute | Details |
|---|---|
| Stack | Python, EDA, statistical visualization |
| Dataset | Mind the Gender Wage Gap |
| Notebook | View on Kaggle |
| Dataset | Description |
|---|---|
| NATO Alliance Dataset | 32 countries, 10,700+ rows of country statistics, military equipment, and NATO operations data |
| GlobalSpace — A Century of Space Missions (1957–2035) | 10,500+ mission records across 12 space agencies/operators, 26 attributes |
| Global Military Arsenal Dataset: Weapons Systems | Structured dataset on global military weapons systems for defense analytics |
| Esports World Cup 2025 Dataset | 27 tournaments, $100M+ prize pool, rosters and club standings from Riyadh 2025 |
Rubixe AI Solutions, Bangalore
August 2025 — February 2026
- Completed a 6-month internship applying data science consulting concepts to real-world business problems, bridging theoretical foundations with practical implementation.
- Delivered consistent, high-quality results across client-facing data science tasks, demonstrating strong technical proficiency and work ethic.
Gandhinagar University
Bachelor of Technology in Artificial Intelligence
July 2023 — August 2027
-0F172A?style=for-the-badge&logo=tableau&logoColor=white&labelColor=2563EB){kind=icon}
Learning:
- Advanced Machine Learning Engineering
- Data Science for Real-World Decision Systems
- Deep Learning Internals
- Speech AI and Audio Intelligence
- MLOps and Production ML Deployment
- Cloud-Native AI Applications
Building:
- Real-Time Sentiment Analysis Pipelines
- Predictive Analytics Platforms
- Insurance Interest Prediction Systems
- Real Estate Valuation Models
- Retrieval-Augmented Q&A Applications
- Recommendation Engines
- Dockerized FastAPI ML Services
Exploring:
- Speech Technologies
- Applied NLP
- AI-Driven Cybersecurity
- Threat Classification
- Anomaly Detection
- Vector Search and RAG Pipelines
- Model Monitoring and Experiment Tracking
Open To:
- Machine Learning Engineer Roles
- Data Science Roles
- AI/ML Internships
- Speech AI Research Opportunities
- MLOps and Model Deployment Roles
- Open Source AI Projects
Turning data into models, and models into usable, deployable systems.