Default profile banner
SK

Sourav Karmakar

@user.2599119

Data Scientist at Sravathi AI Technology Pvt Ltd

Bangalore, India

https://github.com

Sravathi AI Technology Pvt LtdRamakrishna Mission Vivekananda Educational and Research Institute

Sourav Karmakar is a Data Scientist at Sravathi AI Technology with expertise in graph neural networks, deep learning, NLP, and MLOps. He has tackled graph generation problems using Graph Attention Networks, implemented LSH combined with FAISS Vector Index databases for 20x acceleration in embedding retrieval, and developed robust AWS data pipelines. He holds an M.Sc. in Big Data Analytics from Ramakrishna Mission Vivekananda Educational and Research Institute and has qualified GATE in Data Science & AI.

Experience

Data Scientist

Sravathi AI Technology Pvt Ltd

Full-time•Aug 2022 - Present•Bangalore, India

Tackled Graph Generation problem using SoTA Graph Attention Network improving top-5 accuracy by 13%. Achieved 7-point RMSE reduction using XGBoost for reaction yield prediction. Implemented LSH with FAISS for 20x reduction in similar data fetching time. Deployed and maintained backends of 3 production projects using Docker, Django, Git. Implemented AWS data pipeline using S3, EMR, Redshift, Sagemaker.

Machine Learning Intern

Crediwatch Information Analytics Pvt Ltd

Full-time•Feb 2022 - Jun 2022•Bangalore, India

Classified 50M+ News Articles into 3 categories with TF-IDF and XGBoost. Improved F1-Score from 0.96 to 0.99. Developed in-house MLOPs platform for model building. Implemented and tracked Data Drift and Concept Drift of deployed models.

Education

Ramakrishna Mission Vivekananda Educational and Research Institute

M.Sc.

Big Data Analytics

Jan 2020 - Jan 2022

Banwarilal Bhalotia College

B.Sc.

Mathematics

Jan 2017 - Jan 2020

Licenses & Certifications

Certification on Generative AI with Large Language Models

• No expiration

Skills

Python
R
NumPy
Pandas
Scikit-learn
Statsmodels
SciPy
PyTorch
PyG
Multiprocessing
TensorFlow
LangChain
MLFlow
NLTK
OpenCV
RDKit
OpenAI
FAISS
Machine Learning
Deep Learning
XGBoost
Graph Neural Network
NLP
KNN
SVM
Random Forest
LSTM
CNN
Transformers
BERT
LLM
Generative AI
AWS S3
AWS EMR
AWS Redshift
AWS Sagemaker
Databricks
Docker
Django
Linux
Git
MySQL
PostgreSQL
MongoDB