About

Learn more about me

Master's in Computer Science Student

AI/ML Engineer with 4+ years of experience designing, developing, and deploying scalable AI and machine learning solutions. Proficient in deep learning, NLP, generative AI, predictive modeling, and advanced analytics. Skilled in data engineering, cloud platforms, MLOps, and model optimization with a focus on regulatory compliance, model explainability, and operational efficiency. Adept at leveraging AI/ML to drive actionable insights, automate workflows, and enhance business decision-making across diverse domains.

  • Birthday: September 5th 2000
  • Phone: +1 (937) 313-3022
  • City: Ohio, United States.
  • Age: 25
  • Degree: Master's in Computer Science
  • Email: ketan.joshi590@gmail.com

Master’s student in Computer Science specializing in data analytics and AI/ML, with expertise in automation for process optimization. Experienced in developing predictive models and optimizing workflows to enhance efficiency. Proficient in Python, SQL, and Power BI, with a strong ability to lead cross-functional teams and deliver data-driven solutions.

Projects

Languages

Internship

Courses

Skills

Python 80%
SQL 75%
HTML & CSS85%
Java 70%
Power BI 75%
AI/ML 85%
Automation 75%
c/c++ 70%

Software and Technologies

Visual Studio Code

Jupyter Notebook

TensorFlow

OpenCV

Django

pandas

Amazon Web Services - AWS

MongoDB

SQL

excel

Power BI

Google Cloud Platform

Tableau

Android Studio

MATLAB

Git Hub

Interests

Reading

Games

Music

Travelling

Resume

Check My Resume

My Professional Experience

AI/ML Engineer

Aug 2024 — Present

  • Executed AI-powered AML transaction monitoring project using graph machine learning and anomaly detection to identify suspicious patterns, increasing true-positive alerts by 27%, reducing false alerts by 34%, and accelerating investigation turnaround across systems.
  • Engineered AI-driven fraud prevention project applying deep learning on card and digital payments, decreasing fraud losses by 22%, improving detection precision by 19%, and enabling near-real-time risk scoring for customers across global platforms.
  • Designed AI-powered credit decisioning project leveraging alternative data and gradient boosting models, increasing approval accuracy by 16%, reducing bias metrics by 12%, and shortening loan decision times across channels for retail banking users.
  • Established AI-based virtual assistant project using generative transformers for customer support, improving resolution accuracy by 24%, reducing average handle time by 31%, and increasing self-service adoption across digital banking platforms globally at scale.
  • Enhanced AI-powered market risk analytics project using time-series forecasting and deep neural networks, improving volatility prediction accuracy by 21%, lowering stress-test computation time by 28%, and supporting trading risk decisions across capital markets.
  • Automated AI-driven document intelligence project transforming KYC and onboarding workflows using NLP and OCR, reducing manual review effort by 38%, improving processing turnaround by 26%, and ensuring regulatory compliance enterprise-wide for financial institutions.

    • Company - JPMorgan Chase & Co.
    • Location - Remote, USA

    AI/ML Engineer

    Jan 2021 — Aug 2023

  • Developed a smart CRM and SaaS analytics platform leveraging AI/ML to analyze customer behavior, subscription trends, and product engagement using SQL, Python, Zoho Analytics, and Azure Synapse, improving strategic alignment by 15%.
  • Designed scalable ETL pipelines using Zoho Creator, Python, SQL, and Azure Data Factory, performing feature engineering, data cleaning, transformation, and integration from CRM, billing, and support systems, enhancing processing efficiency and analytics delivery by 12%.
  • Built predictive subscription churn and upsell models using XGBoost and LightGBM, training models on historical usage and engagement data, documented methodology, and collaborated with product teams, improving forecast accuracy and retention strategy effectiveness by 18%.
  • Optimized predictive models through hyperparameter tuning and cross-validation, integrating customer tenure and usage features, maintaining detailed documentation and sharing results with stakeholders, improving model reliability and business decision accuracy by 16%.
  • Conducted advanced customer segmentation, cohort analysis, and lifecycle modeling using K-Means and DBSCAN, evaluating with precision, recall, and F1-score, integrating Azure Machine Learning pipelines, increasing retention by 12% and repeat subscription adoption by 10%, collaborating with marketing and analytics teams.
  • Developed interactive dashboards and executive reports using Zoho Analytics and Power BI to visualize revenue, adoption, engagement, and support metrics, reducing manual reporting by 14% and accelerating data-driven decision-making, with thorough documentation for team use.
  • Optimized SaaS business operations by integrating AI/ML insights into CRM workflows, automating recommendation engines, supporting product management decisions, and collaborating with DevOps for model deployment, retraining pipelines, drift monitoring, and CI/CD integration, enhancing efficiency and engagement by 15%.

    • Company - Zoho Business Services LLP
    • Location - MH, India

    My Education

    Master of Science in Computer Science

    Aug 2023 — May 2025.
    GPA: 3.58/4.00

    I am currently persuing Master's in Computer Science from the University of Dayton at Dayton, Ohio.

    • Institute - University of Dayton.
    • Location - Dayton, Ohio United States.
    • Key Coursework: Database Management, Data Structures, Deep Learning, Algorithm Design, System Analysis

    Bachelor of Engineering in Computer Science

    Aug 2018 — June 2022
    GPA: 3.30/4.00

    I have completed my Bachelor's Degeer in Computer Science from The Savitribai Phule Pune University (SPPU), Pune.

    • Institute - All India Shivaji Memorial Society's COE.
    • Location - Pune, Maharashtra, India.
    • Key Coursework: Distributed Systems, Data Analytics, Cloud Computing, AI & ML, Data Mining and Warehousing

    Higher Secondary Science Stream

    June 2016 — Jul 2018

    I finished my Higher Secondary(11th & 12th) from Vidya Mandir Prashala and Jr. College, Miraj. I completed my Higher Secondary's in Science Stream.

    • Institute -Vidya Mandir Prashala and Jr. College
    • Location - Miraj, Maharashtra, India.

    Recent Certifications

    Google Advanced Data Analytics - Professional Certificate

    March 2025

    Microsoft Power BI Data Analyst Associate

    March 2025

    Projects

    Walmart Sales Forecasting & BI Dashboard

    Analyzed 400K sales records using Python, SQL, and Power BI to build time-series forecasting models achieving 87% accuracy, uncovering a 20% holiday spike, and delivering actionable business insights through an interactive BI dashboard.

    Movie Recommendation System

    Developed a web-based movie recommendation system using Python, JavaScript, and HTML/CSS, implementing cosine similarity and fuzzy matching, creating a typo-tolerant, personalized search UI that boosted demo user engagement by 30% among classmates and faculty.

    Portfolio

    My Work

    • All
    • Python
    • Data Analysis
    • Deep Learning
    • Web
    • MATLAB

    Real-Time Object Detection System

    A real-time object detection model using YOLO v5, optimized for mobile devices, identifies household items like t-shirts, balls, and vacuum cleaners. This project enables efficient object detection for smart home automation and inventory management, balancing speed and accuracy for seamless mobile deployment in practical applications.

    Movie Recommendation System

    This Python-based movie recommendation system suggests similar movies based on user input using content-based filtering. It calculates similarity scores from features like genres, cast, and keywords using cosine similarity. Enter your favorite movie, and it recommends the top 10 closest matches. Easily customizable for data and features.

    Walmart Sales Analysis

    This project analyzes Walmart’s weekly sales (2010–2012) using Python, SQL, and Power BI. It explores seasonality, holiday effects, CPI, and unemployment trends. ML models predict sales, ARIMA & SARIMAX forecast trends, and K-Means & Random Forest segment stores to identify key sales drivers through interactive dashboards & SQL insights.

    Medical_Chatbot

    An AI-powered medical chatbot built with Python, TensorFlow, and NLP, capable of recognizing 200+ diseases with 90% accuracy. Supports English, Hindi, and Marathi, with audio output for accessibility. Optimized with data augmentation & hyperparameter tuning, reducing response time by 50%.

    8 Puzzle Problem

    This project implements Iterative Deepening Search (IDS) and A Algorithm* to solve the 8-Puzzle Problem efficiently. The solution explores state-space search strategies, leveraging depth-first search with iterative deepening and heuristic-based informed search.

    Diwali Sales Data Analysis

    This project analyzes Diwali sales data to identify key trends and patterns. It employs data cleaning, exploratory data analysis (EDA), and visualization techniques to uncover actionable insights that can drive future sales strategies.

    F1 Race Management

    A comprehensive database management system for Formula 1 racing, handling entities like Races, Drivers, Teams, Results, and Seasons. This system enables efficient data storage, historical record-keeping, and analysis, supporting Formula 1 analytics and reporting for enhanced insights into race data and performance

    Image Processing with Histogram Matching and Filtering

    This project implements custom MATLAB solutions for histogram matching and image filtering (median and adaptive) without built-in functions. Perfect for enhancing images, reducing noise, and experimenting with advanced image processing techniques!

    GRE words desktop notification using python

    This Python script sends desktop notifications with random GRE vocabulary words and their meanings using the plyer library. It reads from a JSON file and sends notifications at regular intervals (default: 30 minutes). Ideal for GRE preparation or vocabulary building. Customizable for notification timing and icon.

    Contact

    Contact Me

    My Address

    348 Firwood Drive Apt. A, Dayton 45419, Ohio, United States.

    Social Profiles

    Email Me

    ketan.joshi590@gmail.com

    Call Me

    +1 937 313 3022

    Designed by Ketan Joshi