Scribd Logo
Upload

Welcome to Scribd!

  • 2 views

    Libin Kutty Resume - pdf-1

    Uploaded by

    Prasad Supare

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Libin Kutty Resume - pdf-1

    Uploaded by

    Prasad Supare
    2 views2 pages

    Original Title

    Libin Kutty Resume.pdf-1

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    2 views2 pages

    Libin Kutty Resume - pdf-1

    Uploaded by

    Prasad Supare

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Libin Kutty DATA SCIENTIST

    Magdeburg, Germany
     (+49) 17673550378 |  [email protected] |  libinjohn26 |  libin-kutty

    Skills
    Programming Languages Python, HTML, CSS, SQL, Java(basics)
    Frameworks Pytorch, Tensorflow, Streamlit, Keras, openCV
    Packages NumPy, Pandas, SciPy, Matplotlib, Sklearn, Transformers
    DL Techniques NLP, Graph Neural Network, Computer Vision, 3D Mesh, BERT, CNNs, LLMs, Knowledge Graph
    Other Skills Information Retrieval(IR), Explainable AI(XAI), Git, Data Parsing, Jupyter Notebooks
    Languages English (Bilingual, C1), German (Beginner, A1), Hindi & Malayalam (Native)

    Experience
    Syenah GmbH Eschborn, Germany
    DATA SCIENTIST March’23 - April’23
    • Domain: Deep Learning, NLP, Data & Visual Analytics, Web Scraping
    • Stack: Python, Streamlit, Transformers, OpenAI, NumPy, Pandas, Matplotlib
    • Worked with large language models like GPT3 and GPT3.5 using prompt engineering to extract entities from the text
    • Created a demo application to show and visualize the results
    • Worked with web scraping tool to extract Twitter information and elastic search to update the index

    ifak- Institut für Automation und Kommunikation Magdeburg, Germany


    RESEARCH ASSITANT March’20 - Feb’23, May’23 - Present
    • Domain: Requirement Engineering, Machine Learning, Deep Learning, NLP, BERT, Semantic Roles
    • Stack: Python, PyTorch, NumPy, Pandas, Matplotlib, Scikit-learn, Transformers, Allen-NLP, SpaCy, NLTK, Streamlit
    • Developed a natural language processing algorithm to translate behavioral requirements to a logical structure, achieved 90% entity
    accuracy and 55% exact match
    • Created a demo application using Streamlit to train and evaluate machine learning models
    • Developed new algorithms using part of speech and dependency tags, and Semantic role labeling for requirement formalization
    • Worked on data annotation on requirements-related categories to train a NER model
    • Worked on web crawling applications to collect dataset
    • Published the new algorithms and results in conferences
    • Researched and created a new algorithm with continual learning and different learning approach as part of my master thesis for
    requirement classification

    Otto-von-Guericke University - AI LAB Magdeburg, Germany


    RESEARCH ASSISTANT Nov’20 - Mar’22
    • Domain: Scheduling, Deep Learning, Graph Neural Network
    • Stack: Python, PyTorch, PyTorch-Geometric, Numpy, Pandas, Matplotlib
    • Developed an algorithm for machine job shop scheduling problems with the help of Graph Neural Network
    • The algorithm is scalable and capable of handling large volumes of data
    • Developed different deep learning architectures to explore and evaluate model performance
    • Created a synthetic dataset to facilitate model training and enable effective performance evaluation
    • Implemented a reinforcement learning approach with graph neural network

    Vistex Asia-Pacific Private Limited Mumbai, India


    ASSOCIATE SAP DEVELOPER July’16 - Mar’19
    • Domain: Software Engineering, Databases, SAP, UI
    • Stack: SAP ABAP
    • Utilized ABAP programming language to implement new functionalities and create applications
    • Improved the user interface application and managed the database related to each application
    • Lead the entire application development process, from initial planning to delivering to maintenance

    Education
    Otto-von-Guericke Universität (OvGU) Magdeburg, Germany
    MASTER OF SCIENCE IN DATA AND KNOWLEDGE ENGINEERING SPECIALIZING IN DATA SCIENCE AND NLP (GRADE:
    2.0/5.0)
    • Thesis Topic: Continual Lifelong Learning for NLP-related Requirement Classification (Grade: 1.7)
    - Utilization of different learning paradigms and integration of knowledge graph

    Rizvi College of Engineering, Mumbai University Mumbai, India


    BACHELOR IN ENGINEERING IN COMPUTER ENGINEERING FIELD
    • Thesis Topic: Facial Recognition using Neural Network
    Academic Projects
    Automated pre-processing of 3D aneurysm meshes

    • Domain: Computer Vision, 3D Mesh, Machine Learning, Deep Learning, Graph Neural Network, Medicine
    • Stack: Python, PyTorch, PyTorch-Geometric, NumPy, SciPy, Matplotlib, Blender
    • Aimed at focusing on designing a neural network to perform pre-processing of 3D aneurysm meshes like mesh cutting with segmen-
    tation
    • Created a new graph neural network architecture to process 3D mesh data
    • Overcame the laborious and time-exhausting manual step of pre-processing by achieving 80% mean IoU

    Legal norm retrieval and Legal textual entailment classification

    • Domain Deep Learning, Information Retrieval, Entailment, BERT, Legal


    • Stack: Python, PyTorch, Sklearn, NumPy
    • Aimed at exploring and evaluating legal document retrieval and entailment technologies as a part of COLIEE competition
    • Implemented different techniques such as TF-IDF, sentence BERT, Legal BERT, and Graph neural network
    • Achieved the highest F1 score in legal retrieval task among 18 different submission

    Deep Learning based Anomaly Detection using Ensembles with Leave-out classes

    • Domain Deep Learning, Natural Language Processing, Computer Vision, Ensemble


    • Stack: Python, NLTK, Tensorflow
    • Aimed at detecting anomalies in a given dataset (image: MNIST, CIFAR, and text: 20 Newsgroups) by exploiting the softmax distribution
    of predictions from an ensemble of classifiers, each leaving a class out.
    • Implemented new and multiple statistical approaches to evaluate the softmax distribution
    • Achieved better results than baseline for MNIST by 7% and comparable results for other datasets mentioned before

    IRTEX: Image Retrieval with Textual Explanations

    • Domain Deep Learning, Machine Learning, Explainable AI, Computer Vision, Information Retrieval
    • Stack: Python, Tensorflow, openCV, Pandas, NumPy, SciPy, Matplotlib
    • Aimed at developing an image retrieval system with explainable AI to help users understand the results
    • Enabled user to understand the workings of an Image Search Engine by providing visual and textual explanations consisting of Global
    (refers to the whole result set based on the query) & Local (individual explanations for each retrieved image) explanations.
    • Consolidated that explanations are useful and required through user acceptance of 85% and 65%, respectively, in a user study.
    • Concluded that explanations provide transparency via high-level features (semantic and segmentation) with an 80% majority.

    Scientific Publication
    • Natural Language Processing for Requirements Formalization: How to Derive New Approaches? - Book Chapter - Concurrency,
    Specification, and Programming
    • Using Textbook Knowledge for Statute Retrieval and Entailment Classification - New Frontiers in Artificial Intelligence: JSAI-isAI
    2022 Workshop, JURISIN 2022, and JSAI 2022 International Session
    • Geometric Deep Learning for Aneurysm Diagnosis - CMBE’21: 7th International Conference on Computational and Mathematical
    Biomedical Engineering, June 27–29, 2022, Vol 2 Page 699
    • Automatisierte Modell- und Testgenerierung aus textuellen Anforderungen - Book: Automation 2022 Publisher: VDI Verlag, Page
    521
    • Automated Requirement Formalization using Product Design Specifications - REFSQ’22: 5th Workshop on Natural Language
    Processing for Requirements Engineering (NLP4RE), March 21, 2022
    • IRTEX - Image Retrieval and Textual Explanation - IEEE 2nd International Conference on Human-Machine Systems (ICHMS), Sept.
    8-10, 2021
    • Legal Norm Retrieval with Variations of the BERT Model Combined with TF-IDF Vectorization - ICAIL ’21: International Conference
    on Artificial Intelligence and Law, June 21–25, 2021
    • Applying BERT Embeddings to Predict Legal Textual Entailment - Journal - The Review of Socionetwork Strategies, February 19,
    2022

    You might also like