Data Science Intern - ML/NLP Engineer - Python & TensorFlow Specialist

Mrinmoy Nanda

Bengaluru, Karnataka +91-892******* *******.*********@*****.***

LinkedIn GitHub

PROFESSIONAL SUMMARY

Data Science Intern with hands-on experience building and deploying end-to-end ML and NLP solutions across healthcare, recommendation systems, and text classification. Skilled in Python, scikit-learn, and TensorFlow, with strengths in feature engineering, model evaluation, hyperparameter tuning, and interactive application deployment. Proven ability to build scalable ML pipelines, improve model performance, and deliver production-ready analytical solutions. TECHNICAL SKILLS

• Programming Languages: Python, SQL

• Machine Learning & Deep Learning: scikit-learn, TensorFlow, PyTorch, XGBoost, CNN, RNN, LSTM, BiLSTM

• NLP: TF-IDF, Word2Vec, Transformers, NLTK, spaCy, Text Classification, Sentiment Analysis

• GenAI & LLMs: Hugging Face, LangChain, LlamaIndex, OpenAI API

• Data Handling & Visualization Tools: Pandas, NumPy, Matplotlib, Power BI, PostgreSQL, Excel, Streamlit

• Concepts: Feature Engineering, Model Evaluation, Hyperparameter Tuning, Cross-Validation PROFESSIONAL EXPERIENCE

Data Science Intern May 2025 – February 2026

Ai Variant, Bengaluru, Karnataka, India

• Engineered and productionized 3+ end-to-end ML/NLP applications using Streamlit for interactive inference and user- facing demos.

• Processed and transformed structured and unstructured data, enhancing feature quality and boosting model perfor- mance by 15%.

• Built and tuned classification models using Logistic Regression, SVM, and BiLSTM, achieving up to 94.3% macro F1-score through cross-validation and hyperparameter tuning.

• Designed reusable data-processing and prediction pipelines, reducing inference latency by 30%.

• Evaluated models using Precision, Recall, F1-score to ensure robust performance on imbalanced datasets.

• Collaborated with cross-functional teams to translate business requirements into AI-driven solutions and improve op- erational efficiency.

PROJECTS

Human Emotion Detection System GitHub Live Demo Nov 2025 – Feb 2026 Tech Stack: Python, BiLSTM, TensorFlow, Keras, Streamlit

• Developed a multiclass NLP system classifying 6 human emotions using Linear SVM and BiLSTM models.

• Improved minority class recall to 0.86 by handling class imbalance using focal loss, class weighting, and dropout.

• Achieved 83% macro F1-score, delivering balanced performance across emotion categories.

• Deployed an interactive Streamlit application for user input and live prediction. Drug Reviews Condition Classifier GitHub Live Demo Aug 2025 – Nov 2025 Tech Stack: Python, TF-IDF, Logistic Regression, scikit-learn, Streamlit

• Engineered an NLP classification pipeline using TF-IDF and Logistic Regression to categorize drug reviews into 3 medical conditions.

• Enhanced prediction reliability through preprocessing, confidence-threshold filtering, and sentiment/side-effect analysis.

• Attained 94.3% macro F1-score on real-world healthcare data.

• Productionized the solution as an interactive Streamlit application for end-user usage. Book Recommendation Engine GitHub Live Demo May 2025 – Aug 2025 Tech Stack: Python, Pandas, Collaborative Filtering, Cosine Similarity, Streamlit

• Designed a collaborative filtering recommendation engine using user-item interaction matrix and cosine similarity.

• Reduced dataset sparsity by 20% through optimized filtering techniques.

• Generated personalized top-N recommendations to enhance user relevance.

• Delivered an interactive Streamlit-based recommendation interface. EDUCATION

PG Diploma in Data Science September 2024 – Present ExcelR Edtech Pvt Ltd

Bachelor of Technology in Information Technology CGPA: 8.3 July 2020 – June 2024 Maulana Abul Kalam Azad University of Technology, West Bengal 1

CERTIFICATIONS & TRAINING

Data Science Certification – ExcelR Data Analytics Training – Accenture Data Analytics Masterclass – Udemy Power BI – Simplilearn Customer Segmentation & Consumer Behaviour – Udemy MS Excel – Simplilearn 2

Contact this candidate