J.KANISHKHA

Masters in Data Science

August 2024 - May 2026(Expected)

Relevant Coursework

• Introduction to Data Science
• Probability and Statistics for Data Science
• Programming for Data Science

Bachelors in Computer Science

August 2019 - July 2023

Relevant Coursework

• Data Structures and Algorithms
• Machine Learning
• Deep Learning
• Data Science

Xneuronz AI

March 2024 - Present

Machine Learning Engineer

• Built a Floor Plan search model using OpenAI CLIP embeddings and Qdrant vector database.
• Finetuned Stable Diffusion model using QLoRA and LoRA techniques on custom-created Interior Room Design datasets.
• Developed a Multi-Agent System using CrewAI, where AI agents collaboratively strategize room allocation, room placements by following user constraints and design floor layout plans.
• Built a RAG system for the Bangalore Building Bye-Law document using DSPy, LangChain and Unstructured.io.
• Tech Stack: Langchain, LlamaIndex, Stable Diffusion, Huggingface, DSPy, Unstructured.io

Dotlas

September 2023 - January 2024

Data Science Intern

• Built a data product analyzing sentiments in restaurant reviews, identifying impact factors over a timeperiod.
• Used data imputation techniques and advanced ML algorithms to reduce missing values in Real-Estate data by 38%.
• Applied Semi Supervised Learning Techniques to predict house prices and restaurant meal prices.
• Actively engaged in developing production notebooks in Databricks, working on ML pipelines and contributing to data product development.
• Proficiently collected Data from various Restaurant and Retail websites using custom-built collectors using Swarm framework.
• Designed and implemented data processing pipelines using Databricks platform, automating data collection procedures.
• Created parsers within the Honeycomb framework to clean and structure collected data, optimizing it for query operations.
• Executed SQL queries to extract valuable insights and performed data visualization for better decision-making.
• Tech Stack: Python, Jupyter, Databricks, Git, Pandas, SQL, Spacy, NLTK, Scikit-Learn

KPTAC Technologies

February 2023 - July 2023

Data Science Intern

• Scraped major e-grocery websites in UAE with Scrapy, Beautiful Soup, and Selenium, then preprocessed the data.
• Designed and implemented spiders, data pipelines for automated data extraction, parsing and storing using Scrapy.
• Performed Data cleaning and preprocessing using EDA, FE, PCA to ensure accuracy and consistency of scraped data.
• Conducted data analysis and clustering to identify top-selling products & gain customer segmentation insights.
• Used Folium and GeoPandas to optimize customer order locations, achieving a 30% reduction in delivery times.
• Tech Stack: Scrapy, Postgres, BeautifulSoup, Pandas, Numpy, Seaborn, PowerBi, Qlik, Git

Sentient Labs FZ LLC

June 2021 - August 2021

Software Engineering Intern

• Developed a Robot application using ROS in AWS RoboMaker for obstacle avoidance and path planning.
• Tested the robot application with Turtlebot in a simulation environment using AWS S3 bucket, Gazebo, Rviz.
• Deployed the application to an edge device using AWS Greengrass IOT and established a ROS pipeline using AWS.
• Containerized the robot application with Docker for improved portability and scalability.
• Tech Stack: ROS, Docker, Linux, AWS Robomaker, AWS IOT GreenGrass, AWS S3, AWS EC2, Arduino

An End-to-End Hybrid Learning model for detection of Covid-19 from Chest X-ray images

May 2022 - January 2023

Under Prof. Dr. Pranav M Pawar

• Developed a cutting-edge Hybrid Learning Model using a CNN+LSTM approach, achieving 97.56% accuracy in COVID-19 detection from chest X-ray images.
• The study compares the Hybrid Learning Model's performance against established pre-trained models such as VGG19, Xception, and MobileNet.
• Incorporated techniques such as CLAHE Image normalization, SMOTE, HOG, CNN, LSTM, Transfer Learning, and Fine-Tuning.
• Research work was presented and published at International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1), Bangalore, India, 21-22 April 2023.
• doi: 10.1109/ICAIA57370.2023.10169832
• Citatition: K. Jaisankar, P. M. Pawar and D. S. Joseph, "An End to End Hybrid Learning Model for Covid-19 Detection from Chest X-ray Images," 2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1), Bangalore, India, 2023, pp. 1-6, doi: 10.1109/ICAIA57370.2023.10169832.
• Tech Stack: Python, Scikit-Learn, TensorFlow, OpenCV, Matplotlib, Pandas

Classification of Microstructure Images of Metals Using Transfer Learning

Jun 2022 - December 2022

Under Prof. Dr. Angel Arul Jothi

• With light optical microscopes, microstructure images of four different metals were acquired for this task, including copper, mild steel, aluminum, and stainless steel.
• The proposed work employs Fine-tuned transfer learning models namely VGG16, VGG19, ResNet50, DenseNet121, DenseNet169 and DenseNet201 to train and classify the images in the acquired dataset into different classes of metals.
• Research work was presented and published at The International Conference on Modelling and Development of Intelligent Systems (MDIS 2022) Sibiu, Romania, 28-30 October 2022.
• doi: 10.1007/978-3-031-27034-5
• Citatition: Khan, M.A.H., Kanishkha, J., Sabnis, H., Jothi, J.A.A., Prasad, A.M.D. (2023). Classification of Microstructure Images of Metals Using Transfer Learning. In: Simian, D., Stoica, L.F. (eds) Modelling and Development of Intelligent Systems. MDIS 2022. Communications in Computer and Information Science, vol 1761. Springer, Cham. https://doi.org/10.1007/978-3-031-27034-5_9
• Tech Stack: Python, TensorFlow, OpenCV, Matplotlib, Pandas

Application of DCNN for Visual Tracking of Mobile Robots using UAV

August 2022 - January 2023

Under Prof. Dr. Kalaiselvi Venkatesan

• Created custom datasets of Turtlebot2i robot, annotated and augmented the dataset using Roboflow.
• Developed a Deep CNN model using PyTorch to detect the mobile robot by training the model with custom datasets.
• Performed comparative study with Faster-RCNN, SSD, YOLOv5 and YOLOv7 for deployment into dji Tello drone.
• Used PID controller and OpenCV to autonomously track and follow the mobile robot using dji Tello drone.
• LAB: Mechatronics Lab (MS2)
• Tech Stack: Python, Anaconda, PyTorch, OpenCV, Roboflow, TelloSDK

Multi-Modal approach for Autonomous Driving

July 2022 - February 2023

Under Prof. Dr. Pranav M Pawar, Dr. Raja Muthalagu

• Created custom datasets of steering angle, views, and trajectory of the car by driving the car in Udacity self-driving car Simulator.
• Collected datasets from German Traffic Sign Benchmark, KITTI, and Lyft self-driving car open source datasets.
• Developed Multiple Deep Learning Models to detect and classify Traffic signals, detect obstacles and detect lanes.
• Performed comparative study with Mask-RCNN, ResNet50, InceptionV3 and MobileNet in simulated environments.
• Worked on KITTI 3D data visualization, FCNN, DeepSort, MTAN, SFA 3D, UNetXST, Vision Transformer models.
• Built an Autonomous driving vehicle using Jetson Nano, Arduino, and Ultrasonic Sensor that can perform Lane Detection, Obstacle Avoidance and response to Traffic Signals using Deep Learning and Image Segmentation.
• Research work was submitted to Multimedia Tools and Applications, An International Journal, Springer 2023 and is currently under review.
• LAB: Innovation Lab
• Tech Stack: Python, Anaconda, PyTorch, Tensorflow, OpenCV, Udacity self-driving car Simulator

British Airways

November 2023 - December 2023

Data Science Virtual Internship Program

• Conducted web scraping of airline review data from Skytrax, focusing on reviews specifically about the airline itself, utilizing Python in Jupyter Notebook for efficient data collection.
• Employed data cleaning techniques to address the messy and text-heavy dataset, preparing it for analysis.
• Applied Python for in-depth analysis, utilizing topic modelling, sentiment analysis, and word clouds to derive valuable insights.
• Trained a machine learning model, RandomForest algorithm, to predict the target outcome of customer booking.
• Presented findings through a comprehensive PowerPoint slide, incorporating visualizations, metrics, and clear explanations to communicate key analysis results to the management team effectively.
• Skills: Web Scraping, Data Manipulation, Data Visualization, Machine learning, PowerPoint

KPMG

September 2023 - September 2023

Data Analytics Consulting Virtual Internship

• Led a thorough assessment of data quality and completeness for Sprocket Central Pty Ltd's datasets, ensuring accuracy for subsequent analysis.
• Developed a strategic customer targeting approach, focusing on high-value segments, through a three-week analysis covering Data Exploration, Model Development, and Interpretation phases.
• Engineered a Tableau dashboard to visually present key data summaries and strategic insights, aiding in marketing and growth strategy formulation.
• Skills: Data Quality Assessment, Data Insights, Data Insights and Presentation

Cognizant

September 2023 - September 2023

Artificial Intelligence Virtual Experience Program

• Executed a comprehensive data science workflow encompassing exploratory data analysis, data modelling, machine learning production, and performance evaluation using Python.
• Developed and communicated a plan to address the client's problem statement, including a detailed data model diagram and presentation slide for effective leadership communication.
• Applied industry-standard practices such as CRISP-DM, data cleaning, feature engineering, and data visualization, showcasing proficiency in Python, communication, quality assurance, and model interpretability.
• Skills: Exploratory Data Analysis, Data Modeling, Model Building and Interpretation, Machine Learning, Production Quality Assurance

BCG X

August 2023 - September 2023

Data Science Virtual Experience Program

• Conducted thorough EDA on historical customer and pricing data to understand data types, statistics, and variable distributions.
• Verified the hypothesis of price sensitivity by defining and calculating relevant metrics correlated with churn.
• Prepared a concise half-page summary of key EDA findings. Recommended additional data sources from the client and suggested open-source datasets for augmentation.
• Skills: Business Understanding & Hypothesis Framing, Exploratory Data Analysis, Feature Engineering & Modelling, Findings & Recommendations

Languages and Databases

Frameworks

Tools