Kiran Viswanadhapalli

About Kiran Viswanadhapalli

Machine Learning Engineer | AI Systems | Computer Vision | Real-Time Inference | AWS | PyTorch | TensorFlow I build production-ready AI systems that deliver reliable performance in real-world, latency-sensitive environments. My work focuses on taking models beyond experimentation—designing and deploying end-to-end machine learning pipelines that scale. I have hands-on experience across the full ML lifecycle, including data preprocessing, feature engineering, model training, evaluation, inference optimization, and cloud deployment. I’ve developed and deployed systems across: Computer Vision: Object Detection (YOLO, Faster R-CNN), Image Segmentation (U-Net, Attention U-Net) Deep Learning: CNN architectures, transfer learning, fine-tuning Real-Time AI Systems: Low-latency inference and performance optimization Time-Series Modeling: Predictive modeling and evaluation Tech Stack: Python, PyTorch, TensorFlow, OpenCV, Docker, FastAPI, AWS (EC2, S3, SageMaker) I specialize in: Building scalable, production-grade ML systems Optimizing models for latency, throughput, and reliability Deploying containerized AI services in cloud environments Bridging research models with real-world applications

Experience

1. Machine Learning Engineer

   M&T Bank

   Feb 2026 — Present · Clanton, AL, US

   Designed and deployed scalable machine learning models for forecasting and risk analytics, enhancing decision-making. • Developed NLP and LLM-based solutions for automated document processing, improving efficiency in handling financial reports. • Built and optimized end-to-end data pipelines for real-time analytics, leveraging tools like Apache Spark and Kafka.

Education

• University of Alabama at Birmingham

  Master of Science, Computer Science