Summary
Overview
Work History
Education
Skills
Timeline
Generic

Akash Andure

Pune

Summary

Results-driven Machine Learning Engineer with 3.7 years of experience and an expertise in OCR pipelines, data extraction, and predictive modeling. Proficient in building and optimizing models using Python, TensorFlow, PyTorch, and Scikit-learn. Adept at data visualization, business intelligence, and cross-functional collaboration, delivering high-performance solutions ahead of schedule. Experienced in deep learning models, NLP, and LLM for advanced data extraction.

Overview

3
3
years of professional experience

Work History

Machine Learning Engineer

Konsultera Solutions Pvt Ltd
02.2023 - Current
  • Led the development of an end-to-end OCR pipeline, managing a team of interns and annotators, to extract data from health insurance policy documents, including forms, tables, and text, standardizing the output into JSON format.
  • Assumed responsibility for client-facing interactions, typically handled by a business analyst, to assess and refine extraction requirements, ensuring precise alignment of data outputs with business needs.
  • Achieved 98% accuracy by developing and optimizing a classification model using EfficientNet, differentiating between useful and non-useful pages within PDF documents, significantly improving data extraction performance.
  • Implemented the Donut model for OCR-free data extraction, automating the identification of key policy details, and improving overall data accuracy.
  • Tested and validated the solution across local and server environments using Postman, ensuring robust performance and functionality for large-scale data processing.
  • Increased efficiency by reducing document processing time to under 5 seconds per document, with an average extraction accuracy rate of 93%, optimizing workflows for high-volume health insurance documents.
  • Developed and implemented a custom splitter module to break down extensive PDF files into manageable sections, enhancing both efficiency and precision in data extraction.
  • Authored and fine-tuned custom prompts for a Large Language Model (LLM) to accurately extract detailed specifications and performance data from complex technical datasheets, streamlining data processing in industrial applications.
  • Automated the generation of JSON files and image annotations for patient medical reports, reducing project timelines from 2-3 weeks to 1 week using regex, fuzzywuzzy, and PaddleOCR.

AI ML Developer

Dual Globe Solutions Pvt Ltd
05.2021 - 02.2023
  • Contributed to data preprocessing, cleaning, and validation processes to ensure dataset integrity and quality for accurate model training and prediction.
  • Performed extensive Exploratory Data Analysis (EDA) using Python libraries (Pandas, NumPy, Matplotlib) to uncover key insights and patterns in customer behavior, guiding model development and feature engineering.
  • Developed and optimized machine learning models, including Logistic Regression, Decision Tree, Random Forest, and Adaboost, achieving high accuracy and efficiency to predict customer EMI delinquency.
  • Collaborated with cross-functional teams, effectively translating complex technical concepts into actionable insights for non-technical stakeholders, enabling informed decision-making.
  • Implemented precise customer segmentation, reducing Turn Around Time (TAT) in customer sourcing by 20%.
  • Participated in the Restaurant Review Sentiment Analysis project, applying NLP techniques such as tokenization, normalization, stop-word removal, lemmatization, and stemming.
  • Built a statistical sentiment classification model using Count Vectorizer, TF-IDF, and Word2Vec, effectively categorizing restaurant reviews as positive or negative sentiments.
  • Enabled the client to proactively engage with customers and gather actionable feedback by providing clear sentiment analysis results, facilitating improvements to customer service strategies.

Education

Bachelor of Engineering -

Savitribai Phule Pune University

Skills

Technical Skills

  • Machine Learning
  • Deep Learning
  • Natural Language Processing (NLP)
  • Large Language Models (LLM)
  • Optical Character Recognition (OCR)
  • Generative AI
  • Data Analytics
  • Business Intelligence
  • Predictive Modeling
  • Data Visualization (Power BI, Matplotlib, Seaborn, Plotly)

Programming Languages

  • Python (PyTorch, Scikit-learn, NumPy, Pandas, OpenCV, Fuzzywuzzy)
  • SQL
  • Regular Expressions (Regex)

Tools & Platforms

  • Azure
  • Postman
  • Power BI
  • Excel
  • Version Control (Git, GitHub, Bitbucket)
  • Jupyter Notebook, Jupyter Lab
  • VS Code

Software Development Skills

  • Object-Oriented Programming (OOP)
  • Operating Systems (OS)

Operating Systems

  • Windows
  • Linux

Soft Skills

  • Cross-functional Collaboration
  • Project Management
  • Problem-Solving
  • Time Management

Timeline

Machine Learning Engineer

Konsultera Solutions Pvt Ltd
02.2023 - Current

AI ML Developer

Dual Globe Solutions Pvt Ltd
05.2021 - 02.2023

Bachelor of Engineering -

Savitribai Phule Pune University

Similar Profiles

CREATE PROFILE
Akash Andure