Surviving MLOps: a curated map of the sprawl

What it does

awesome-mlops is a hand-curated index of MLOps resources—books, papers, courses, articles, tools, and communities—sorted into roughly twenty categories from core concepts to model governance. It acts as a table of contents for a field that sits awkwardly between data science, software engineering, and infrastructure. There is no code here, just links and brief citations.

The interesting bit

The list looks past the usual pipeline tools. It catalogs sections on product management, the economics of ML, and “People & Processes,” acknowledging that production ML fails as often on org charts as on tensor shapes. That breadth is rare in a discipline usually obsessed with YAML.

Key highlights

• Covers the full lifecycle: training, deployment, monitoring, feature stores, and DataOps.
• Includes non-technical corners like ML product management, responsible AI, and team structure.
• Curated by Dr. Larysa Visengeriyeva, with references to standards such as the MLOps SIG specification and Google’s Practitioners Guide.
• Points to free courses (MLOps Zoomcamp) alongside O’Reilly books and academic papers.
• Links to active communities (MLOps.community, DataTalks.Club) for when the documentation inevitably contradicts itself.

Caveats

• Every resource is an outbound link with no archival fallback, so longevity depends entirely on external hosts.
• The README is built from nested <details> blocks; browsing without Ctrl+F is tedious.
• Curation context varies widely: some books get full author, year, and publisher metadata, while other entries are little more than a title and a URL.

Verdict

Grab this if you are an ML engineer, platform lead, or technical PM who needs a crash-course syllabus or interview prep. Skip it if you are looking for a framework to clone; this is a library card catalog, not a codebase.