Summary
Overview
Work History
Education
Skills
Accomplishments
Timeline
Generic

Abhi Vishrut Minocha

Bengalore,Karnatka

Summary

Data Scientist familiar with gathering, cleaning and organizing data for use by technical and non-technical personnel with experience in Machine Learning and Deep Learning.. Highly organized, motivated and diligent with significant background in NLP.

Overview

4
4
years of professional experience

Work History

Data Scientist - 2

Dun & Bradstreet
04.2022 - Current
  • Built financial Information chat bot using Open-CV and OCR techniques to extract text from PDF document.
  • Used Langchain to split text into chunks and create embeddings to store in Vector DB.
  • Then passed these embeddings as context to LLM model.
  • User query was answered using passed context.
  • Used semantic similarity approach to filter out news related to natural calamities using SBERT model to be used in ESG scoring model.
  • Built custom Spacy NER model to extract entities from text like "consequences", "lives lost".
  • Streamlined data collection methods to improve quality control measures and minimize errors in analysis results from various sources like news, annual reports, etc.
  • Cleaned transformed, and analyzed large datasets to uncover hidden trends and patterns for actionable insights.

Software Engineer

L&T-NxT (Acquired By Mindtree In July 2021)
08.2020 - 04.2022
  • Built cognitive search application created for site engineers to search results from SOP's.
  • Implemented various preprocessing techniques before passing PDF to model and Elasticsearch to retrieve answers faster.
  • Built platform used for processing structured and unstructured legal manufacturing documents and identification of risky clauses in construction domain.
  • Developed Tensorflow1 based Object-detection document -layout model for NLP-based platform.
  • Fine-tuned BERT model for Contract data and built FLASK/FAST APIs to inference models.
  • Handled end-to-end development and deployment of machine learning as well as deep learning models
  • Converted PyTorch-based BERT Question/Answering model to ONNX model which resulted in faster inference.
  • Handled docker-based deployment and container orchestration using shell scripting.

Education

Diploma - Artificial Intelligence And Machine Learning

University of Hyderabad
Online
12.2022

B.Tech - Information Technology

SRM Institute of Science And Technology
Chennai, India
05.2020

Skills

  • Python Programming
  • Machine Learning
  • SQL Databases
  • Statistical Analysis
  • Scikit-Learn
  • Natural Language Processing
  • Large Language Models
  • Langchain
  • Vector DB
  • Fast/Flask API
  • Pyspark
  • Docker

Accomplishments

  • Received spot-on award for handling the developing and deploying platform end to end.
  • Received Gratitude for creating New Insights using the existing data.

Timeline

Data Scientist - 2

Dun & Bradstreet
04.2022 - Current

Software Engineer

L&T-NxT (Acquired By Mindtree In July 2021)
08.2020 - 04.2022

Diploma - Artificial Intelligence And Machine Learning

University of Hyderabad

B.Tech - Information Technology

SRM Institute of Science And Technology
Abhi Vishrut Minocha