Default profile banner
Chandan SahooCS

Chandan Sahoo

@chandansahoo

Data Scientist

Pune

github.com/ChandanCharchit

Data Dynamics Inc.National Institute of Technology, Bhopal

Chandan Sahoo is a Data Scientist with experience in developing AI solutions and implementing advanced NLP models. He possesses expertise in utilizing BERT for NER tasks, building robust APIs with fastAPI, and leveraging technologies like Streamlit and Apache NiFi. Currently pursuing an M.Tech in Artificial Intelligence, he has hands-on experience with multi-modal similarity search and analyzing complex datasets.

Experience

Associate Data Scientist

Data Dynamics Inc.

Full-timeMay 2021 - PresentPune

• Created streamlit multiapp which helped users manage different functionalities of Data Science Engine of the product. • Implemented BERT for NER tasks of classifying PII data such as Name, Location, Organisation etc. in documents and thus calculating risk scores of documents. • Added service to the app using Apache nifi and Kibana. • Built APIs with fastAPI for various functionalities such as listing, executing and saving entities, patterns, keywords and values. Also created APIs for adding respective configurations to the database. • Worked with Regex and spaCy pattern matching which matched 86 different entities which are classified as risky data. • Added several test cases using python unit testing framework which made code refactoring easy and ensured each component benefited by achieving a quality product. • Gave sessions to the team on best practices to be followed for many of the technologies used in code implementation for the product.

Education

National Institute of Technology, Bhopal

M.Tech. in Artificial Intelligence

Artificial Intelligence

Jan 2020 - PresentGrade: CGPA 8.0

College of Engineering & Technology, BBSR

B.Tech. in Mechanical Engineering

Mechanical Engineering

Jan 2014 - Jan 2018Grade: CGPA 8.16

Licenses & Certifications

Python for Everybody

Coursera

• No expiration

Professional Data Science Certification

IBM

• No expiration

Deep Learning Specialization

Deeplearning.AI

• No expiration

Skills

Python
ScikitLearn
Keras
fastAPI
spaCy
RegEx
unittest
Streamlit
Linux
Docker
Tensorflow
Apache nifi
Snorkel
PostgreSQL
Git