MLOps / DevOps
Production infrastructure, deployment pipelines, and lifecycle management of ML workloads.
About
I am a DevOps Engineer at Vodafone, with a strong interest in machine learning and the engineering behind intelligent systems. My interests span deep learning, language models, and computer vision, with a particular curiosity about mechanistic interpretability— understanding the internal computations of neural networks.
Beyond industry work, I write technical articles and founded Vector Intuition, a website for learning about AI, previously known as CrackSheet. CrackSheet was live until I stopped it a few months ago. Vector Intuition is under development.
Focus
Production infrastructure, deployment pipelines, and lifecycle management of ML workloads.
Predictive modeling, feature engineering, and model evaluation using classical ML methods.
Neural network design for computer vision and NLP, with emphasis on modern architectures and optimization.
Object detection, image segmentation, and real-time visual understanding pipelines.
Language-model applications, retrieval pipelines, and agent-style workflows.
Research interest in reverse-engineering neural networks — circuits, features, and the internal algorithms models learn.
Selected Work
A website for learning about AI.
A tool for managing, summarizing, and interacting with research papers using large language models.
Personalized recommendation engine for books, songs, and movies with collaborative filtering.
Medical imaging classifier for brain tumor detection using convolutional neural networks.
Real-time gesture recognition system for interpreting sign language.
Predictive model for early heart attack risk assessment from clinical data.
Writing
Notes on machine learning, engineering, and the tools I use.