A Temporal Convolutional Network-based Approach for Network Intrusion Detection
R. Nazre, R. Budke, O. Oak, S. Sawant, A. Joshi. (2024). "A Temporal Convolutional Network-based Approach for Network Intrusion Detection." IEEE ICIICS 2024.
CV
Education
- M.S. in Computer Science, University of Massachusetts Amherst, 2027 (Expected)
- B.Tech. in Computer Engineering, College of Engineering Pune, 2025
- Honors in Data Science, CGPA: 8.59/10
Work Experience
- Summer 2024: Research Intern, TCS Research (Pune, India)
- Curated a structured dataset of character relations from unstructured narrative text (Harry Potter corpus).
- Conducted comparative experiments with multilabel classification algorithms (Random Forest, SVM, Neural Networks) using LLM embeddings (BERT, DistilBERT), achieving 89.6% accuracy.
- Designed and implemented a Neo4j-based knowledge graph for analyzing relational structures.
- Technologies: Python, PyTorch, Neo4j
Skills
- Programming Languages: Python, C++, Java, C, JavaScript, R, Julia
- Databases & Web: SQL, MongoDB, Cassandra, Neo4j, HTML/CSS, Node.js, Express, React
- ML Frameworks & Tools: PyTorch, LangChain, Hugging Face, Scikit-learn, TensorFlow, Spark, Kafka, Docker, Linux, LaTeX
Publications
A Novel Multivariate Bi-LSTM Model for Short-Term Equity Price Forecasting
O. Oak, R. Nazre, R. Budke, Y. Mahatekar. (2024). "A Novel Multivariate Bi-LSTM Model for Short-Term Equity Price Forecasting." IEEE GCAT 2024.
A Comparative Analysis of CNN-based Deep Learning Models for Landslide Detection
O. Oak, R. Nazre, S. Naigaonkar, S. Sawant, H. Vaidya. (2024). "A Comparative Analysis of CNN-based Deep Learning Models for Landslide Detection." IEEE ACOIT 2024.
A Novel Transfer Learning based CNN Model for Wildfire Susceptibility Prediction
O. Oak, R. Nazre, S. Naigaonkar, S. Sawant, A. Joshi. (2024). "A Novel Transfer Learning based CNN Model for Wildfire Susceptibility Prediction." IEEE INCET 2024.
Projects
- InsightFlow: Serverless AI Analytics Agent (Dec 2025): Architected a serverless text-to-SQL agent on Cloudflare Workers using Llama 3; engineered a custom linear regression forecasting engine in TypeScript to analyze 1,000+ sales records with <50ms latency.
- AI Interview Companion (RAG Agent) (Dec 2025): Engineered a serverless RAG system using Cloudflare Workers and Vectorize, featuring a custom Python evaluation pipeline that validated 90.3% semantic retrieval accuracy against ground-truth interview data.
- Narrative Sense Disambiguation with LLMs (Dec 2025): Engineered a word-sense plausibility pipeline using DeBERTa/RoBERTa and GPT-4 synthetic data.
- Scalable Sequence Modeling (Mamba vs. Transformers) (Dec 2025): Benchmarked computational complexity and memory scaling of Selective SSMs.
- Reverse-Mode Automatic Differentiation Engine (Oct 2025): Developed a computational graph framework from scratch to enable gradient computation via backpropagation; engineered 11 core operations including matmul and logdet.
- TERRA-CD: Semantic Change Detection (March 2025): Created a benchmark dataset of 5,221 paired Sentinel-2 images; trained Siamese networks for change detection.