A list of my machine learning related publications
Machine Learning Books 📚
I had the unique opportunity to co-author two machine learning books. More to come …
“Machine Learning Production Systems” by O’Reilly Media
In the rapidly evolving landscape of artificial intelligence, building machine learning systems that work reliably in production environments remains one of the greatest challenges facing organizations today. Over the course of 2023 and 2024, I collaborated with Robert Crowe, Emily Caveness, and Di Zhu to create “Machine Learning Production Systems”, published by O’Reilly Media—a comprehensive guide designed to bridge the gap between experimental machine learning and production-ready systems. As the capabilities of generative AI continue to transform industries, we recognized the need to take a fresh look at the ML production landscape. Our goal was to examine the different components that come together to create robust ML production systems, with special attention to the unique requirements of GenAI deployments. “Machine Learning Production Systems: Engineering Machine Learning Models and Pipelines” addresses the critical need for practical guidance by providing a comprehensive roadmap for the entire ML production lifecycle. Drawing from our collective experience at organizations ranging from innovative startups to established tech giants, we’ve distilled proven patterns and architectural blueprints that work in real-world settings. The book is organized into four comprehensive sections that systematically cover all aspects of machine learning engineering:
- Data management frameworks and best practices
- State-of-the-art modeling techniques
- Robust deployment strategies
- Critical operational considerations
Throughout these sections, we explore how to efficiently collect and preprocess data, build high-performance models, design resilient serving architectures, and implement effective monitoring systems—all essential elements for successfully moving machine learning from experimentation to production. Whether you’re a machine learning engineer, data scientist, engineering manager, or technical leader, this book provides the practical knowledge needed to build and maintain ML systems that deliver real business value in production environments.
“Building Machine Learning Pipelines” by O’Reilly Media
Between 2019 and 2020, I co-authored the ML engineering publication “Building Machine Learning Pipelines”, published by O’Reilly Media.
Building Machine Learning Pipelines is an excellent resource for anyone looking to get a comprehensive understanding of the entire machine learning pipeline process. The book covers all stages of the pipeline, from data ingestion to model deployment, and even goes beyond that to include the optimization of model deployments, model privacy, and the design of model feedback loops.
One of the features of this book is the attention it gives to the early and late stages of the pipeline. Too often, books and courses on machine learning focus solely on model training and ignore the crucial steps of data ingestion, preprocessing, as well as model evaluation and deployment. Building Machine Learning Pipelines does a great job of explaining these stages and the importance of getting them right.
The book also covers the model deployment process. It walks readers through the different options available for deploying models, including using cloud platforms and containerization, and provides guidance on choosing the best option for a given situation. It also discusses the importance of monitoring and maintaining deployed models and provides tips on how to do so effectively.
One of the most impressive aspects of Building Machine Learning Pipelines is its coverage of model optimization. This is an often-overlooked topic in machine learning, but it’s critical for getting the most out of deployed models. The book provides a thorough overview of techniques such as A/B testing, and explains how to use them to optimize model performance.
Overall, Building Machine Learning Pipelines is an invaluable resource for anyone looking to understand and build effective machine learning pipelines. Its comprehensive coverage of the entire process, from data ingestion to model optimization, makes it an essential read for anyone looking to take their machine learning skills to the next level.
“NLP in Action” by Manning Publishing
“Natural Language Processing (NLP) in Action” is a comprehensive guide to the exciting field of natural language processing written in 2017. In this book, we took a dive deep into the fundamentals of NLP and explore the techniques and tools used in the industry.
First, we will cover the basics of NLP, including statistical methods such as bag of words and n-grams. These techniques allow us to extract meaningful information from large amounts of text data and are an essential foundation for many NLP tasks.
Next, we will delve into advanced techniques such as Word2Vec, which allows us to represent words as vectors in a continuous space and capture the meaning of words in a more intuitive way. This is a crucial step in many NLP applications, such as language translation and sentiment analysis.
We will also introduce deep learning, a powerful set of techniques that has revolutionized many fields, including NLP. We will cover convolutional and recurrent neural networks, which are commonly used in NLP tasks such as text classification and language modeling.
We will also explore the use of seq2seq networks, which are a type of neural network designed for tasks such as machine translation and text summarization. These networks can learn to map a sequence of input data to a sequence of output data, making them particularly well-suited for NLP tasks.
Finally, we will discuss the training of models on GPUs, which allows us to train large and complex models much faster than on traditional CPUs. This is particularly important in NLP, where the datasets can be very large and the models can be computationally intensive.
By the end of this book, you will have a solid understanding of NLP and the tools and techniques used to build real-world NLP systems. Whether you are a beginner looking to get started in NLP or an experienced practitioner looking to deepen your knowledge, this book is for you. By understanding the details of NLP and neural networks, it will make you appreciate transformers even more.