👨‍💻 Hey there, my name is Kuntal, and I am an MS grad @ASU.

🚀 I have an interest in ML, DL, and if you want to view my projects, make sure to check my pinned projects below.

⚡ Fun fact: I prefer using tabs over spaces.

PinnedLoading

1. Arxiv-Papers-Recommendations-using-RAGArxiv-Papers-Recommendations-using-RAGPublic

   A retrieval-augmented generation (RAG) pipeline using Elasticsearch & FAISS to index 3,000+ ArXiv papers, improving search efficiency by 35%. Deployed on AWS (S3, DynamoDB) with Docker & CI/CD, ens…

   Jupyter Notebook

2. Efficient-Query-Processing-on-Distributed-Systems-Efficient-Query-Processing-on-Distributed-Systems-Public

   Designed a scalable ETL pipeline using Docker, Neo4j, Kafka, Hadoop, and Kubernetes to process 1.5B+ rows (~50GB). Configured Minikube for Kubernetes orchestration, optimizing resource allocation &…

   Python

3. Log-based-System-Anomaly-Detection-Log-based-System-Anomaly-Detection-Public

   Developed a log anomaly detection pipeline for 575K+ log blocks using Drain, Sentence-BERT, and Bi-LSTM, achieving 98.32% F1-score. Automated data preprocessing & model training (80% reduction in m…

   Python

4. Scalable-Data-Ingestion-Query-Optimization-for-Large-Scale-Reddit-DataScalable-Data-Ingestion-Query-Optimization-for-Large-Scale-Reddit-DataPublic

   Built a high-performance ETL pipeline to process 10M+ Reddit comments into PostgreSQL, optimizing ingestion with pg_bulkload (under 300s insertion time). Designed optimized SQL queries for subreddi…

   Shell