Machine Learning Artificial Intelligence

Mahendra Avudiyappan

Chicago, IL Ph: +1-312-***-**** *******************@*****.*** https://github.com/Mavudiya https://www.linkedin.com/in/mahendra-avudiyappan-083a54240/ SUMMARY

AI enthusiast with hands-on experience in building and deploying machine learning solutions in education and healthcare. Skilled in fine-tuning LLMs, optimizing RESTful APIs, and implementing scalable MLOps workflows. Curious and active in exploring emerging AI tools and technologies and consistently committed to upskilling and contributing to real-world impact through scalable AI deployments. EDUCATION

DePaul University – 3.5 GPA Chicago, IL

Master of Science in Artificial Intelligence Jan 2024 – Nov 2025 Course Work: Machine Learning Operations and Computer Vision MVJ College of Engineering – 3.52 GPA Bangalore, Karnataka Bachelor of Engineering in Computer science Sep 2019 – Sep 2023 SKILLS

● Programming Languages: Python, Html5, CSS, JavaScript

● Framework & Libraries: Django, TensorFlow, PyTorch, Scikit-learn, NLTK, NumPy, Pandas

● Database & Data Tools: MySQL, JSON, RESTful APIs (design and testing)

● Cloud & Deployment: Microsoft Azure, MLOps Workflows, CI/CD, API Integration

● Version Control & Dev Tools: Git, GitHub, Bitbucket, Docker, VS Code, Jira

● Productivity & Collaboration: MS Office, Google Workspace, Zoom, Wrike, Slack CERTIFICATION

● AI Engineer – AIROBOSOFT

● Azure Certification – Microsoft

● ML Engineer – Saint Louis University

PROJECTS

LLM-Powered Recommendation Chatbot for Universities Nov 2024 – present

● Led evaluation and fine-tuning of LLaMA 3.2, LLaMA 3.3, and DeepSeek models, and other transformer models to identify and integrate the most effective university recommendation system, reducing recommendation errors by 25%.

● Enhanced response accuracy by 30% using FuzzyWuzzy-based fuzzy matching, structured JSON storage, and machine learning-driven segmentation, improving user satisfaction.

● Developed and deployed a REST API and integrated Retrieval-Augmented Generation (RAG) to enable real-time, context- aware recommendations, resulting in a 50% decrease in response latency.

● Optimized backend efficiency by 40% through advanced data processing, NLP enhancements, and model fine-tuning, ensuring scalability for increasing user queries.

Brain Tumour & Alzheimer detection using Deep learning Dec 2022 – May 2023

● Developed a Convolutional Neural Network (CNN) model for brain tumour detection and classification using MRI scans, achieving 97.36% accuracy through optimized deep learning techniques.

● Enhanced segmentation precision by 20% using UNET architecture, improving tumour localization and classification accuracy through advanced image processing techniques.

● Designed an end-to-end pipeline for automated image pre-processing, augmentation, and classification, improving inference speed by 35%.

● Published research findings in JETIR, validating model efficacy and contributing to peer-reviewed academic literature.

(JETIR Paper Link).

WORK EXPERIENCE

ML Engineer - Saint Louis University. Sep 2023 – Oct 2023

● Independently designed and deployed a machine learning model using Linear Regression, improving budget accuracy by 15% through advanced predictive analytics.

● Examined and modeled historical financial trends with time series analysis, boosting forecast precision by 20% for annual budget planning.

● Leveraged supervised learning techniques to optimize budget allocations, supporting data-driven decision-making for

$100K+ in financial planning.

● Collaborated with cross-functional finance and tech teams to scale AI insights across budget planning systems to incorporate AI-driven insights into the budgeting process, reducing resource allocation errors by 10% and enhancing project planning efficiency.

Contact this candidate