0% found this document useful (0 votes)
2 views1 page

Anshuman.Mandal-profile

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 1

ANSHUMAN MANDAL

7765828832 anshuman.mandal.01@gmail.com Linkedin PortfolioWebsite Github Medium Tableau

WORK EXPERIENCE
EY| Associate Technology Consultant | Data Analytics Service Line July 2023 - Present
Developed and deployed an order management system using Python, SQL, and Azure Services to
support global operations across European, African, and US time zones for Unilever. Utilized Excel,
Databricks, and PowerBI for ETL data processing, creating dashboards and performing analytics for
automation and insights.
Resolved 100+ tickets for 40+ chatbots, focusing on stabilizing bot performance and retraining NLP
solutions. Improved key metrics such as response accuracy by 15%, reduced false positives by 10%, and
enhanced response time by 20%. Led the development of logistic automation software for Indian clients.
Collaborated with the Splunk dashboard team for solution monitoring and provided live support during
critical deployments, including pension drives.
Conducted API testing and evaluation for the banking sector, ensuring system reliability. Utilized tools
like Postman, C#, and Azure Function App for API testing and development.
Skyserve.ai | Internship June 2022 - August2022
Integrate Image Processing modules into the company Source Code.
Establish network connection during early boot.
PROJECTS
Logistics Performance Analysis and Delay Prediction and Consumer segmentation
Analyzed logistics data to identify key factors affecting delivery performance followed by Consumer
Segmentation using algorithms(Random Forest, Gradient Boosting)
Built machine learning models to classify orders, achieving 69.82% test accuracy .
Developed insights to reduce delays and boost profitability, visualized via a Power BI dashboard.
Disaster Detection and Classification Using Text, Images, and Real-Time Twitter Scraping
The project uses text and image classification models to detect disaster-related content, achieving up to
81% accuracy for text and 80% for images (VGGNet).
Text classification models include Logistic Regression, Random Forest, SVM, and RNN (LSTM).
A deployed website allows classification on user-uploaded data or real-time Twitter posts using a live
scraper for disaster detection.
Document Image Denoising Using Convolutional Autoencoders
Used a Convolutional Autoencoder to denoise document images while preserving details.
The autoencoder learns key features through unsupervised learning, effectively removing noise.
The approach shows promising results for tasks like document restoration and OCR preprocessing.

EDUCATION
Bachelor of Technology, Computer Science and Engineering 2019 - 2023
PES University, Bangalore SGPA: 9.00
International Public School, Kanke Road, Ranchi 2017 - 2019
12th Grade, 75.8% (Physics Chemistry Mathematics)
St.Thomas School Dhurwa, Ranchi 2005- 2017
10th grade, 93.5%
ADDITIONAL INFORMATION
Technical Skills: Deep Learning, Machine Learning, Data Analytics, Predictive Modelling, Feature
Engineering, Computer Vision, Recommendation systems, Large Language Models, Data Visualization
Retrieval Augmented Generation, Database Management System, Generative AI, SQL, GitHub, Prompt
engineering, Exploratory Data Analysis (EDA), Clustering, Dimensionality Reduction, Anomaly Detection,
Hypothesis Testing, Regression, Time Series forecasting, Deployment
Tools: Python, C++, C#, Pytorch, Tensorflow, Keras, Streamlit, Llama 2, GPT3.5 Transformers, YOLO, CNN,
Natural Language Toolkit, scikit-learn, OpenCV, Matplotlib MySQL, Power BI, Tableau, MS EXCEL, Azure,
Seaborn, NumPy,Spark, Hadoop, Big Query, Flask
Soft Skills: Startup Vision, Communication, Leadership, Teamwork, Problem Solving, Lead Collaboration

You might also like