I am a results-driven AI/ML Engineer with a B.Tech in Computer Science & Engineering from Trident Academy of Technology (Graduated May 2026, CGPA: 6.8 / 10).
I specialize in translating deep learning theory into production-ready software, bridging Computer Vision, Natural Language Processing (NLP), and Full-Stack Backend Architectures. With hands-on dual-internship experience, I build systems designed to operate under real-world constraints.
- Facial Recognition Attendance System: Developed an end-to-end biometrics pipeline (live video capture → Haar Cascade face detection → CNN face identification → MySQL logging). Adopted for live operational attendance.
- Object Detection Pipelines: Built custom object detectors utilizing transfer learning with state-of-the-art CNN architectures, applying data augmentation (rotation, contrast, scaling) to close the generalization gap.
- Model Lifecycle Management: Handled raw data preprocessing, architecture design, hyperparameter tuning, and strict evaluation metrics (Precision, Recall, F1-Score).
- NLP Classification Pipeline: Designed a multi-class text classification pipeline (tokenization → TF-IDF → Word Embeddings → Deep Neural Network) trained on Kaggle datasets.
- Landmark Detection & Image Classification: Built custom CNN models in TensorFlow/Keras to perform structural landmark mapping and image categorization.
- Robust Generalization: Optimized training workflows using dropout layers, batch normalization, and early stopping to mitigate overfitting.
| Category | Technologies |
|---|---|
| Languages | Python, JavaScript (ES6+), HTML5, CSS3, SQL (MySQL) |
| AI/ML & Deep Learning | TensorFlow, Keras, OpenCV, DeepFace, Dlib (68 Face Landmarks), Haar Cascades |
| Libraries & Data Science | NumPy, Pandas, Matplotlib, Seaborn |
| Backend & Databases | Flask, Flask-WTF, Flask-Session, MySQL |
| Tools & Platforms | Docker, Docker Compose, Git & GitHub, Stripe API, Jupyter Notebooks, Google Colab |
| Evaluation Metrics | Precision, Recall, F1-Score, AUC-ROC, Accuracy |
An advanced, secure, and intelligent online examination platform designed to conduct objective, subjective, and practical coding tests while ensuring academic integrity.
- Proctoring Core: Integrated real-time face tracking and identification using DeepFace and Dlib to verify student identity and prevent academic dishonesty.
- Full-Stack Backend: Built with Flask, securing sessions via Flask-WTF and Flask-Session, backed by a relational MySQL database.
- Integrations & Deployment: Payment processing via Stripe API and containerized using Docker & Docker Compose for zero-configuration deployments.
A live biometrics pipeline that captures video, identifies users, and logs timestamps in a backend database.
- Pipeline: Live Camera Feed ➡️ Haar Cascade Detection ➡️ CNN Verification ➡️ MySQL Logger.
- Achievement: Fully adopted for operational use during the CTTC MSME internship.
An end-to-end NLP pipeline for categorization of multi-class textual corpora.
- Tech Stack: NLTK, TF-IDF Feature Extraction, Custom Word Embeddings, Deep Neural Network (Sequential API) with Dropout and Batch Normalization.
├── index.html # Portfolio Single-Page Application (HTML5, CSS3, JS)
├── Profile.jpeg # Main portfolio section profile image
├── Passport Size.jpg # Favicon tab icon
├── Aditya_Swain_Resume.pdf # Downloadable PDF Resume
└── README.md # Personal profile & project documentation
I am always open to discussing new opportunities, collaborations, or technical challenges in the AI/ML space.
- Email: swainaditya85@gmail.com
- LinkedIn: linkedin.com/in/aditya-ranjan-swain
- GitHub: github.com/Aditya1791