6 questions · pass at 80% to earn the journey badge. You can retake it anytime.
01 Which best distinguishes deep learning within AI?
It is unrelated to machine learning It is machine learning using multi-layer neural networks that learn features automatically It only uses hand-engineered features It is a database technique
02 In a transformer, self-attention primarily enables what?
Sequential one-token-at-a-time processing only Modeling relationships between all tokens in parallel, capturing long-range context Eliminating the need for training data Hard-coded grammar
03 What characterizes an agentic "ReAct"-style loop?
Reason then act: interleaving reasoning steps with tool actions and observations Rendering animations Recompiling the model each step Refusing to use tools
04 For production prompt systems, which matters most?
Maximizing ambiguity Determinism aids: explicit constraints, structured output, evaluation and versioning Never testing prompts Avoiding examples
05 What links sim-to-real transfer in robotics with modern AI?
It is unused in robotics Policies trained in simulation are transferred to real robots, leveraging perception + learned control It only applies to chatbots It removes the need for sensors
06 Which governance posture fits high-stakes enterprise AI?
Ship unsupervised and hope for the best Risk-tiered controls, human oversight, audit logging, and continuous evaluation No documentation Self-certifying models