Manasi Bandhaokar

Business Development Executive

Nagpur, Maharashtra, India4 yrs 11 mos experience
Most Likely To SwitchAI Enabled

Key Highlights

  • Developed ML models for credit risk estimation.
  • Created computer vision model for passenger counting.
  • Conducted ML knowledge sessions for industry relevance.
Stackforce AI infers this person is a Data Science and Machine Learning specialist in the Fintech and Transportation sectors.

Contact

Skills

Core Skills

Machine LearningStatistical AnalysisData ScienceComputer VisionArtificial Intelligence

Other Skills

AlgorithmsArtificial Intelligence (AI)C (Programming Language)C++Convolutional Neural Networks (CNN)Data AnalyticsData AnnotationData CleaningData StructuresData VisualizationDeep LearningDeep Neural Networks (DNN)ExcelFeature SelectionFlask

About

Manasi is an enthusiastic fresher with a Dual Degree specializing in Artificial Intelligence and a minor in Computer Science and Engineering looking for research-oriented industry roles. She has previously interned at Citi in using Deep Learning for Credit Risk Modelling. She has gained experience reading and applying research literature to solve problems via coursework projects and research internships. She is excited to use her knowledge in Machine Learning to challenging business problems. She possesses strong interpersonal and communication skills and is a team player. Programming Languages: C/C++, Python, SQL Tools/Technologies/Platforms: PostgreSQL, Insomnia, Anaconda, Git/Github, Tableau, Excel Libraries / Frameworks: Keras/TensorFlow, Flask, Psycopg, Selenium (Web-Driver), unittest, NLTK , Sklearn, Pandas, Numpy, Matplotlib Relevant Coursework – Machine Learning, Deep Learning, Natural Language Processing, Reinforcement Learning, Generative and Graphical Models, Graph Machine Learning, Big Data Processing, Dependable and Secure AI ML, Algorithms and Data Structures, Computer Architecture and Operating Systems

Experience

Citi

3 roles

Business Analyst II

Aug 2024Present · 1 yr 7 mos

Business Analyst

Jul 2023Present · 2 yrs 8 mos

  • Involved in developing ML models for estimating consumer credit risk.
  • Tech Stack : Python, SAS, Excel, Tableau
  • Machine Learning | Statistical Analysis
PythonSAS (Software)ExcelTableauMachine LearningStatistical Analysis

Data Science Intern

May 2022Jan 2024 · 1 yr 8 mos

  • Worked with the Scoring team of Risk Management Unit to develop a challenger model to the existing acquisition risk prediction method for THD Canada portfolio and obtained a lift of 2% in KS score across development, in-time validation and out of time validation data. Recieved a pre-palcement offer based on my contribution.
  • Cleaned THD Canada’s customer dataset having 704 attributes for 200,000 accounts by treating missing, default and outlying values
  • Removed features with low variance among values or having low Information Value and high correlation to obtain a set of 190 features
  • Experimented with over 200 combinations of neural network hyperparameters and used shapley values to select 60 best features
  • Performed a sanity check on the model by comparing the sign of slope of PDP plots with sign of correlation of features with the bad rate
  • Finalized a neural network with 37 features that achieved a KS score of 68% with average PSI of 0.0023 on out of time validation set
Microsoft ExcelMicrosoft PowerPoint

Dexter squad

Machine Learning Instructor

Feb 2023May 2023 · 3 mos · India · Remote

  • Conducted cohort based knowledge sessions on Machine Learning aimed at imparting industry relevant ML knowledge to participants from basic to advanced. Covered topics from Regression to Deep Learning.
Deep LearningArtificial Intelligence (AI)Microsoft PowerPointMachine LearningData Science

Eat neat project

Back End Developer

Jan 2022May 2022 · 4 mos · India

  • 1. Designing statistical experiments for collecting data to validate of prediction models
  • 2. Design and implementation of API for WOFOST crop yield prediction model
  • 3. Contributed to the development of ETL pipeline
  • 4. Scraping data from various websites and designing databases to store it
FlaskMicrosoft PowerPoint

University of ottawa

NLP Research Intern

Dec 2020Jan 2021 · 1 mo

  • Working on summarization of financial documents using NLP and Deep Learning techniques.
  • Exploring approaches to quantify document similarity
Microsoft PowerPoint

City future lab

Summer Research Intern

Apr 2020Sep 2020 · 5 mos · Kharagpur, West Bengal

  • I worked on the project - Application of Artificial Intelligence in bus transit system. I was responsible for developing a computer vision model for determining the count of passengers waiting on the bus stands from the surveillance feed.
  • I conducted an extensive literature survey of publications in the crowd counting domain to identify efficient and accurate CNN architectures for accomplishing the above objective.
  • I created a custom testing set from Google Images. After manually cleaning the dataset, I used an open-source MATLAB script to annotate the dataset. I used python to preprocess the dataset and obtain density map corresponding to each image.
  • I implemented Multi-Columnar Neural Network from scratch using Pytorch and OpenCV, trained it on Shanghai Tech dataset (Part B) and tested its performance on the custom dataset.
  • Subsequently, I applied CSRNet to observe further improvement in the accuracy. I trained the model on a hybrid dataset using transfer learning. Further fine-tuned the model to improve accuracy.
  • Finally applied S-DCNet and was able to obtain a MAE of 6.6 on the test set using pre-trained weights. The test set had a maximum count of ~800 and minimum count of ~15.
Microsoft PowerPointSeleniumComputer Vision

Vertexplus

Machine Learning Intern

Dec 2019Jan 2020 · 1 mo · Jaipur Area, India

  • VertexPlus Technologies Private Limited is a Global Technology Company delivering excellence across different industry verticals.
  • Built a framework for Recommendation Engine using collaborative filtering
Microsoft PowerPoint

Kharagpur data analytics group

Member

Sep 2019Oct 2020 · 1 yr 1 mo

  • KDAG is a student driven research group aimed at bringing Data Analytics and Machine Learning enthusiasts together under the umbrella of a single society.
Microsoft PowerPoint

Business club, iit kharagpur

2 roles

Subhead

Jul 2019Jan 2020 · 6 mos

  • As a sub-head, I was involved in organizing and conducting knowledge meetings on Data Analytics, Finance and Business Strategy for Associate Members.
Microsoft PowerPoint

Associate

Aug 2018Jul 2019 · 11 mos

  • As an Associate Member, I contributed to the organization of 2019 edition of the Indian Case Challenge - largest case study challenge in Asia.
  • Was also actively involved in Knowledge Meets conducted on topics from Data Analytics, Finance and Business Strategy.
  • Business Club, popularly known as B-Club, is IIT Kharagpur’s in-campus society, under the Technology Students'​ Gymkhana, for all things business. It is a knowledge sharing interface for like-minded individuals operating through a ‘knowledge-session’ approach. The Club also reaches out to the student community by organizing Open House sessions, and increasingly, the general public, through initiatives like the Financial Literacy Drive
Microsoft PowerPoint

Education

Indian Institute of Technology, Kharagpur

Dual Degree — Civil Engineering / Artificial Intelligence Machine Learning and Applications

Jan 2018Jan 2023

Dr. Ambedkar College, Nagpur

Subjects

Jan 2015Jan 2017

Modern School , NEERI

Jan 2015Present

Stackforce found 100+ more professionals with Machine Learning & Statistical Analysis

Explore similar profiles based on matching skills and experience