Preetham's Photo

Preetham Ganesh

Machine Learning Engineer

About Me

"Creativity is seeing the same thing but thinking differently"- Dr. A.P.J Abdul Kalam.

Hello, I'm Preetham Ganesh, an ML Engineer. I'm passionate about building and deploying efficient Machine Learning applications to solve real-world problems. I thrive on seeking new challenges to continuously improve my skills and expertise.

I hold an MS in Computer Science from the University of Texas at Arlington, where I worked at the Vision-Learning-Mining Research lab, developing an ASL-to-English translation application, which fueled my passion for inclusive technology. I later joined Qualitest, supporting Google as a Data Analyst to work on the analysis of results for queries submitted to Google Search Engine. Later, as an ML Engineer at Clarium, I worked on custom OCR and NLP applications to extract and summarize information from healthcare and financial documents.

I'm excited to connect with professionals, share insights, and explore new opportunities. If you're interested in collaborating, discussing exciting projects, or simply connecting over shared interests, please feel free to reach out at [email protected].

Experience

March 2022 - March 2024

Machine Learning Engineer @ Clarium

Designed & deployed advanced ML models, including TensorFlow-based Segmentation & Word Recognition models as part of Health & Medical Document Summarization tool. Optimized model selection with MLFlow, fine-tuned the Google T5-Small LLM, and boost model performance with synthetic data augmentation.

  • TensorFlow
  • Natural Language Processing (NLP)
  • Computer Vision
  • MLFlow
  • Micro Services
  • OpenCV
  • SQL
  • Docker
  • Keras
  • REST APIs

October 2021 - March 2022

Data Analyst @ Qualitest

Enhanced data quality by 15% with SQL optimizations and created dashboards for improved decision-making. Automated data workflows with Python, reducing manual tasks by 10%, and ensured data consistency across diverse datasets through cross-functional collaboration.

  • Google BigQuery
  • Data Analysis
  • SQL
  • Knowledge Graphs
  • Python

February 2020 - May 2021

Developed a cascaded ML model system to convert ASL videos into English speech in real-time. Migrated OpenPose from OpenCV to PyTorch, reducing extraction time by 90% and enabling real-time analysis. Optimized pipeline performance through hyper-parameter tuning, achieving a 98% Top-5 accuracy on the WLASL benchmark for Video Sign Language Recognition.

  • TensorFlow
  • OpenCV
  • PyTorch
  • Computer Vision
  • Natural Language Processing (NLP)
  • Academic Publishing

March 2018 - April 2019

Undergraduate Student Researcher @ Amrita Vishwa Vidyapeetham

Developed hybrid ensemble models for rainfall prediction in Tamil Nadu, utilizing bagging and boosting techniques. Created district-specific and cluster-based models to address regional differences, achieving over 91.13% accuracy in forecasting.

  • Regression
  • Scikit-Learn
  • Matplotlib
  • SciPy
  • NumPy
  • Pandas

Skills

Programming Languages
  • Python
  • R
  • SQL
  • MySQL
  • BigQuery
  • MATLAB
  • HTML
  • CSS
  • JavaScript
  • Java
Cloud & DevOps
  • AWS EC2
  • AWS S3
  • AWS SageMaker
  • AWS ECR
  • AWS ECS
  • Azure ML
  • GCP
  • Heroku
  • Docker
  • Git
  • GitHub
  • TensorFlow Serving
  • GitHub Actions
  • GitLab CI
  • CI/CD
Packages & Frameworks
  • TensorFlow
  • Keras
  • Scikit-Learn
  • PyTorch
  • MLFlow
  • NLTK
  • SpaCy
  • NumPy
  • OpenCV
  • Flask
  • Multiprocessing
Data Visualization Tools
  • Pandas
  • PySpark
  • Power BI
  • Tableau
  • Hadoop
  • Scala
  • Snowflake

Publications

2021

POS-Tagging based Neural Machine Translation System for European Languages using Transformers

Authors: Preetham Ganesh, Bharat S. Rawal, Alexander Peter, Andi Giri

This study addresses language barriers by proposing a novel Neural Machine Translation (NMT) approach using inter-language word similarity and Part-of-Speech (POS) tagging for model training and testing. Two classical architectures, Luong Attention-based Sequence-to-Sequence and Transformer models, were used, with tokenization by SentencePiece and Subword Text Encoder, respectively. The models were evaluated on Spanish, French, and German datasets with BLEU, Precision, and METEOR scores, showing promising results.

2020

Personalized system for human gym activity recognition using an RGB camera

This paper presents a Human Activity Recognition system using an RGB camera to classify gym activities (e.g., push-up, squat) through models like SVM, Decision Tree, KNN, and Random Forest, with the latter achieving 98.98% accuracy. A repetition counter was developed using local minima analysis and dynamic time warping to assess workout accuracy per skeletal point. An interactive Android app was also built to provide users insights into their workouts.

Education

August 2019 - May 2021

Master of Science in Computer Science @ University of Texas at Arlington
  • Computer Vision
  • Special Topics in Intelligent Systems
  • Machine Learning
  • Neural Networks
  • Data Mining

July 2015 - April 2019

Bachelor of Technology in Computer Science & Technology @ Amrita Vishwa Vidyapeetham
  • Intelligent Systems
  • Natural Language Processing
  • Software Engineering
  • Database Management System