A timed high-level design interview with an AI interviewer. Talk it through or type, draw on a canvas that simulates your design under load and failure, and get a report that quotes what you said and what your design did.
It is a practice interview run by an AI interviewer instead of a person. On ArchiDepth the interviewer gives you a system design problem, asks clarifying and follow-up questions through each phase, watches your design on the canvas, and keeps time. Afterwards an evaluator scores the session and cites the moments it scored.
The report gives an overall score out of 100, a score out of five for each dimension, and a level estimate with a likely range and a confidence. Each score links to its evidence: a quote from your transcript or a result from simulating your design. A single point of failure under the fault scenario caps reliability.
Yes. You can choose a spoken interview and talk it through with a voice interviewer, or type every answer. Captions and typing always work, so a microphone problem never ends the session. You check your microphone before the interview starts, and you can delete your transcript afterwards.
Requirements, estimation, design and deep dive on a timed plan; wrap-up always starts two minutes before the end.
The final design is simulated, so the report can say what saturated and what failed instead of guessing.
Each dimension cites quotes from your transcript and results from your design, so you know what to fix.
Recordings and transcripts are yours to delete; the scores stay so your progress still adds up.
Start with the free course, build your first system on the canvas in a free practice problem, and see how it holds under load.
