System Design Path is the architecture track in the MathPath family. The public object is a structured design prompt with a visible rubric. A strong answer has to state assumptions, quantify load, draw boundaries, model data, handle failure, instrument the system, and defend tradeoffs.
The ML system design track adds the pieces ordinary backend prompts often miss: feature lineage, offline and online evaluation, model rollout, drift, training / serving skew, and cost at serving time.
The initial public set has eight classic system design prompts and six ML system design prompts. The pages are reviewed, canonicalized on systemdesignpath.com, and included in the live sitemap.