A passionate Software Engineer turned Data Scientist with close to 15 years of work experience.

Education

• 2023 - Masters in Business Analytics from NUS Singapore (4.4 GPA).

• 2007 - Bachelors in Computer Science Engineering (73%).

Work Experience

• 2024-Present
  • ST Engineering Senior CyberSecurity AI Scientist
    • Worked on AI security research with an objective to make AI models more secure and Robust.
    • Devised a novel defense approach called Code Mutation based Triggers Detector (CMTD) to detect backdoor triggers in poisoned datasets and models (specifically Code LLMs).
    • Published a research paper on it.
    • Devised a novel AI explainability based approach to analyze Jail Break Attacks on Large Vision Language Models (LVLMs) which accepts both image and text as inputs.
• 2022-2023
  • Aboitiz Data Innovation Data Scientist
    • Worked on Speech Emotion Recognition(SER) problem and developed a Multi-Modal (text and speech) based model for detecting Emotions and Personality from Raw audio snippets.
    • Published a paper on it.
    • Won the Outstanding Capstone project award at MBS event orgaized by NUS.
    • Tech Stack : Generative AI, Deep learning, Speech processing, PyTorch, DataBricks, Huggingface, Python, Streamlit and others.
  • NUS Computing NLP Fellow
    • Worked in NUS Computing as NLP Fellow.
    • Developed music based language learning material in form of Lyrics Generator project involving Generative AI.
    • Published a paper in ACM Journal.
    • Tech Stack : Deep learning, NLP keyword extraction, finetuning Transformer model, Generative AI, PyTorch, python and others.
• 2021-2022 - Textify.AI Senior NLP Data Scientist
  • Worked as Senior NLP Data Scientist in a startup called Textify.AI.
  • Worked on developing writing and reading assistant application using cutting edge machine learning and deep learning (NLP) technologies.
  • Worked on developing features such as:
    1. Precitive Writing achieved by prompt based finetuning of GPT Neo LLM.
    2. Grammar Corrector achieved by finetuning T5 LLM model.
    3. Essay Grader using deep learning techniques.
    4. Reverse Dictionary using deep learning involving LSTM and RNNs.
  • Tech Stack : PyTorch, AWS, FastAPI,Python, Github, GenAI.
• 2016-2021 - CSG International Associate Solutions Architect
  • Worked as an Associate Architect in CSG Internatioanl Systems.
  • Role involved development on CSG’s singleview product for large scale telecom operators like MTN South Africa, Three UK, MTC Namibia.
  • Played role of scrum master, architect and developer.
  • Largely worked on C, Python,Linux scripting,Perl, and Java.
• 2013-2016 - Oracle Corporation Senior Consultant
  • Worked as a Senior Consultant in Oracle Corp.
  • Role involved development of billing and analytics softwate for large telecom clients.
  • Worked on Machine learning prototype use cases for Telcos such as Churn prediction, Invoice amount anomaly detection, rate plan optimization.
• 2010-2013 - Ness Technologies Senior Software Engineer
  • Worked as a Senior Software Engineer in Ness Technologies.
  • Role involved development and supporting of Oracle Biling and Revenue Management product.
  • Writing code in C, C++, Python, Shell, Perl to implement requirements for Telecom Billing clients and handling Production support.
• 2008-2010 - Mahindra Satyam Senior Software Engineer
  • Worked as a Senior Software Engineer in Mahindra Satyam.
  • Role involved development on Oracle Biling and Revenue Management suite.
  • Writing code in C, C++, Python, Shell, Perl to implement requirements for Telecom Billing clients.

Key Projects

1. Area - Deep learning and Speech processing:
   • MultiModal Speech and Text based Emotions and Personality Recognizer.
   • Speaker Diarization and Audio file Separator
2. Area - Deep learning and NLP:
   • QA Chatbot Gen AI project
   • Predictive Writing Gen AI project
   • Document Analyzer Gen AI project
   • Grammar Corrector
   • Essay Grader
   • Reverse Dictionary
   • Lyrics Generator Gen AI project
3. Area - Deep learning and CV:
   • Automatic detection of formula one cars.
4. Area - Deep Reinforcement learning:
   • Real-time Stock timing model
5. Area - Recommender systems:
   • STEAM games recommender
6. Area - Time Series forecasting:
   • Bitcoin Price Predictor
7. Area - Machine learning based:
   • AirBnB usecase
   • LendingClub bank dataset usecase
   • Telecom Fraud Detection
   • Telecom billing invoice anomaly detector
8. Area - Big Data Spark based:
   • arXiv-searching-TFIDF
   • Amazon reviews finder
9. Area - Python related:
   • Piano music performance analyzer
10. Area - Telecom Billing related:
    • Kafka based stream data processor.
    • Production scripts.

Published Papers

• KeYric: Unsupervised Keywords Extraction and Expansion from Music for Coherent Lyrics Generation

• Speech and Text Based MultiModal Emotion Recognizer

• Steam Games Big Data Recommendation System

Main Accomplishments

➢ More than a decade overall experience, more than 5 years in Data science.
➢ Winner of Coveted Outstanding Capstone project award at a ceremony held at Marina Bay Sands, Singapore
➢ Industry experience in NLP and Generative AI, training LLMs like GPT2, GPT-neo and other GenAI models.
➢ Co-authored and submitted research papers:

• “KeYric: Unsupervised Keywords Extraction and Expansion from Music for Coherent Lyric Generation” ACM Multimedia Conference in 2023.
• “Steam Recommendation System using Pyspark and Big data platforms” ArXiv in 2023.

➢ NUS-Huawei Innovation Challenge (Finalists) – selected as top 5 teams out of 104 total teams (across Singapore).
➢ Experience in working as part of many end-to-end large-scale data-intensive client projects involving production deployments.
➢ Kaggle 1X Expert Tier (In top 1%)..
➢ Exposure to working in 6 different geographies across the world at client site in a consulting role.

Contributory articles/posts

1. Huawei AI Competition
2. Convolutions in CNN
3. Predictive Writing
4. LLM technical aspects
5. Big Data Processing using Pyspark