AI Architect Assessments
Scenario-driven, architect-level questions with full rationale per answer. Free, open to everyone — no login required.
LLM Architecture Fundamentals
Attention, KV cache, decoding strategies, and the failure modes that only show up at production scale.
RAG & Vector Search Systems
Retrieval architectures, embedding choices, ANN index trade-offs, and the failure modes hidden in real corpora.
MLOps & Production AI
Deployment patterns, monitoring, drift, governance, and the on-call reality of running ML in production.
AI System Design at Scale
Distributed training, multi-region inference, capacity planning, and the architectural decisions that hold up at billion-request scale.
Responsible AI, Safety & Governance
Risk frameworks, evaluation, red-teaming, privacy, and the architectural patterns that make governance load-bearing.
Distributed Training & GPU Infrastructure
Parallelism strategies, collective communication, checkpointing, and the GPU/network economics that decide what's actually trainable.
Agentic Systems & Tool Use
Planning loops, tool schemas, memory, evaluation, and the architectural patterns that contain agent failure modes.
Recommender Systems & Ranking
Candidate generation, ranking, calibration, position bias, and the eval methodology that separates real recsys teams from cosine-similarity demos.
Vector DB Internals
ANN algorithms, quantization, hybrid search, and the operational reality behind production vector databases.
AI Cost Optimization
GPU economics, inference cost levers, caching, quantization, distillation, and the architectural moves that reduce cost without lying about quality.
Tip
These are calibrated to a Staff/Architect bar. Choose the best answer, not just a correct-sounding one — every option here has been chosen to look plausible. Read the rationale even on questions you get right.