Machine Learning at Scale: Weekly ML Systems Newsletter
A weekly Substack newsletter unpacking how top tech companies build machine learning systems, written to help ML engineers level up fast.
Overview
Machine Learning at Scale is a Substack publication authored by Ludovico Bessi, a Machine Learning engineer at Google, focused on demystifying how large-scale ML systems are actually built and operated in industry. Each week, subscribers get deep, practitioner-level insights into topics like RAG systems, LLM optimizations, LLM training pipelines, and ML system design—content aimed squarely at engineers who want to move beyond tutorials and understand production-grade ML infrastructure. Rather than surface-level explainers, the newsletter draws on real experience from top tech companies to break down the tools and techniques that separate senior ML engineers from the rest. Beyond the core newsletter, the publication is building out a broader knowledge hub, including an archive of past deep-dive articles, a curated list of tools used by working ML engineers, and upcoming resources like a dedicated ML System Design course and a YouTube channel. It's positioned as a long-term learning companion for anyone serious about mastering the engineering side of machine learning, rather than just the modeling or research side.
Capabilities & Features
- Machine Learning
- ML Systems
- ML Engineering
- LLMs
- RAG Systems
- LLM Optimization
- LLM Training
- ML System Design
- Substack
- Newsletter
Core Features
- Weekly newsletter packed with practitioner-level ML insights
- In-depth deep dives into specific ML engineering topics
- Curated roundups of tools used by real ML engineers
- Searchable archive of past newsletter issues
- Upcoming ML System Design course
- Upcoming YouTube channel for video content
Use Cases
- Upskilling from a general ML practitioner into a systems-focused ML engineer
- Learning how RAG systems and LLMs are optimized and trained at scale
- Discovering new tools and workflows used by industry ML engineers
- Studying ML system design principles ahead of technical interviews
- Staying current on emerging trends in production ML systems
Best For
- Machine Learning Engineers
- Data Scientists
- AI Researchers
- Software Engineers transitioning into ML
- MLOps practitioners
Pros
- •Written by a practicing Google ML engineer with real-world credibility
- •Focuses on practical, production-relevant topics like RAG and LLM optimization rather than theory
- •Consistent weekly cadence keeps readers continuously learning
- •Growing ecosystem of resources beyond the newsletter, including a future course and video content
- •Archive makes it easy to catch up on past insights at any time
Cons
- •Some flagship resources, like the course and YouTube channel, are not yet available
- •Weekly format may be too slow for readers wanting an all-at-once, comprehensive resource
- •Content depth may assume some existing ML background, making it less beginner-friendly
- •Pricing model isn't clearly stated upfront, requiring a visit to the site to confirm
How to Use
Head to the Machine Learning at Scale Substack page and subscribe with your email to start receiving weekly issues directly in your inbox. Browse the existing archive to catch up on past deep dives and tool roundups, and keep an eye out for the upcoming ML System Design course and YouTube channel as they launch.
Frequently Asked Questions
Open in AI Studio
Ask our AI to evaluate if Machine learning at scale fits your specific workflow.
Pricing
Pricing isn't explicitly listed, but as a Substack publication it likely follows a free and/or paid subscription model—check the site directly for current tiers.
Pricing data is provided as a summary. Visit the vendor website for full tier details.