Vivek Senthil
AI/ML Engineer | Full-Stack Developer
M.S. in Artificial Intelligence (RIT) | Specialized in Transformers, LSTMs, and Scalable ML Systems
AI/ML Engineer | Full-Stack Developer
M.S. in Artificial Intelligence (RIT) | Specialized in Transformers, LSTMs, and Scalable ML Systems
I'm an AI/ML Engineer with 3+ years of experience building scalable machine learning systems and full-stack applications. Currently pursuing an M.S. in Artificial Intelligence at Rochester Institute of Technology, with a passion for implementing cutting-edge AI solutions.
My expertise spans Transformer models, LSTMs, time series forecasting (ARIMA/SARIMA), and microservice architectures. I'm experienced in developing end-to-end ML pipelines, optimizing Apache Solr systems, and implementing distributed computing solutions on GPU-accelerated platforms.
I'm particularly drawn to roles that combine AI research with practical engineering, where I can work on challenging problems like natural language processing, computer vision, and predictive modeling for real-world applications.
PyTorch • Transformers • LoRA • Whisper • ASR
MS Capstone Project (Advisor: Zhiqiang Tao) - Parameter-efficient fine-tuning of OpenAI Whisper using Low-Rank Adaptation (LoRA) to improve transcription accuracy for body-worn camera (BWC) footage with law enforcement-specific language and terminology.
R • scikit-learn • ARIMA • LSTM • TensorFlow
Forecasted hourly electricity demand by implementing and comparing classical and deep learning time series models (ARIMA, LSTMs) to capture complex temporal dependencies in 3.5 years of real-world data.
React • Node.js • Spotify Annoy Algorithm
Developed a scalable image search web application for fetching similar products using the Spotify Annoy algorithm for fast approximate nearest neighbor searches.
Python • NLP • Scikit-learn • Logistic Regression
Developed an NLP-based classification system to predict answer relevance in question-answering systems using Logistic Regression and text feature engineering.
CGPA: 3.75/4.0
Coursework: Forecasting Methods, Advanced Computer Vision, High Performance Data Science
CGPA: 9.69/10
CGPA: 8.76/10
"Towards AI-Driven Policing: Interdisciplinary Knowledge Discovery from Police Body-Worn Camera Footage"
Tucson, Arizona | Research Presentation
Presented groundbreaking research on multimodal AI for analyzing police body-worn camera footage to support transparency, accountability, and responsible decision-making in law enforcement. Engaged with researchers and professionals from private companies and police departments across the U.S.
"One-Shot Generative Fake News Detection with Retrieval-Augmented Verification"
Rochester Institute of Technology | Innovation Festival
Showcased scalable, explainable fake news detection system combining structured fact extraction, real-time knowledge retrieval, and LLM reasoning (DeepSeek-r1) with RoBERTa-based contradiction detection. Presented to distinguished visitors including Dr. Cecilia O. Alm and Dr. William H. Sanders.
"Multimodal Analysis of Police Body-Worn Camera Footage Using AI for Behavioral Insights and Accountability"
Rochester Institute of Technology | Graduate Showcase
Developed ML and NLP pipeline with teammates Anita Srbinovska and Angela Srbinovska to analyze police-citizen interactions from body-worn camera footage. Work supported transparency, accountability, and data-driven improvements in law enforcement. Mentored by Dr. Cecilia O. Alm and Dr. Ernest Fokoue.
Mayo Wristband Gesture Recognition AI
NSF Research Traineeship | Spring 2025 | 🏆 Winner
Won the AWARE-AI NSF research traineeship Spring 2025 hackathon! Collected gesture data using Mayo wristband, trained AI model, and successfully deployed it to control PC. Hands-on experience applying machine learning to real-world human-computer interactions.
I'm actively seeking AI/ML Engineer and Data Scientist roles. Feel free to reach out!