Skip to content
All paths
Staff and Architect

Design and run AI systems at company scale.

For senior engineers moving to staff: infrastructure, system design, evaluation strategy, cost, debugging and incident response.

Start the first lessonSee pricing

66 lessons in 4 milestones. Starts in the Systems plan; the full path needs the Mastery plan.

By the end you can
  • Design AI systems and defend the trade-offs in review
  • Set evaluation strategy for a whole product area
  • Own cost, capacity and reliability
  • Lead incident response and postmortems

The roadmap

Work through it in order, or jump to what you need. Open a milestone to see its lessons.

  1. Systems depth

    Systems plan

    Distributed systems and GPU inference you can reason about under load.

    Start: Partitioning & Sharding
    18 lessons
    1. Partitioning & Sharding
    2. Replication & Consistency
    3. Time, Clocks & Causality
    4. Consensus & Coordination
    5. Distributed Transactions & Commit
    6. Queues, Logs & Backpressure
    7. Failure Detection, Gossip & Anti-Entropy
    8. Fault Tolerance & Resilience
    9. CAP, PACELC & Consistency Models
    10. GPU Architecture & the Roofline
    11. VRAM & the GPU Memory Model
    12. Prefill vs Decode
    13. The KV Cache
    14. Attention Cost & Context Scaling
    15. Batching & Continuous Batching
    16. Quantization
    17. Multi-GPU Parallelism
    18. Throughput, Latency & $/Token Economics

  2. Architecture

    Mastery plan

    Infrastructure and system design at scale.

    Start: Docker & Containerization for AI
    15 lessons
    1. Docker & Containerization for AI
    2. Kubernetes & Orchestration
    3. GPU Serving (vLLM / TGI)
    4. Kafka & Event Streaming for Agents
    5. Redis at Scale
    6. OpenTelemetry & Tracing Agent Runs
    7. CI/CD & Autoscaling
    8. Requirements & Constraints
    9. Retrieval Architecture at Scale
    10. Latency Budgets & the Critical Path
    11. Multi-Agent Orchestration Design
    12. Eval & Safety Planning
    13. Reliability, Degradation & Failure Design
    14. Cost Modeling at 10×
    15. The Design Review

  3. Quality and cost

    Mastery plan

    Evaluation strategy and the economics of AI systems.

    Start: Golden Datasets & Labeling
    16 lessons
    1. Golden Datasets & Labeling
    2. Metrics & Scoring
    3. LLM-as-Judge & Bias Control
    4. RAG & Agent Evaluation
    5. Offline vs Online Eval
    6. Statistical Power & Significance
    7. CI Eval Gates
    8. Drift Detection
    9. Unit Economics & $/Request
    10. Token Budgeting
    11. Cost Attribution & FinOps
    12. Infra Budgeting & Margin Analysis
    13. Pricing & Monetization
    14. Scaling Cost Curves
    15. ROI & Build-vs-Buy
    16. Enterprise Procurement Realities

  4. Operate it

    Mastery plan

    Debugging, failure analysis and production incidents.

    Start: The Debugging Method & Root-Cause Analysis
    17 lessons
    1. The Debugging Method & Root-Cause Analysis
    2. Tracing Distributed Failures
    3. Diagnosing Hallucinations
    4. Retrieval Misses
    5. Race Conditions & Deadlocks
    6. Token Explosion & Runaway Loops
    7. Queue Overload & Backpressure
    8. Memory Leaks & Inference Bottlenecks
    9. Provider Outage
    10. Vector DB Corruption
    11. Prompt Injection Breach
    12. Runaway Agent Billing
    13. Queue Backlog Collapse
    14. Latency Cascade
    15. Multi-Agent Deadlock
    16. Hallucinated Finance Response
    17. PII Leakage Incident

  5. Final project

    Ship a real system on real infrastructure. Each project has an architecture diagram and a step-by-step build guide.

    Self-Host & Optimize InferenceAt your volume, hosted-API cost is killing margins. Can self-hosting win, and do you understand why?Architect & Defend a Real System DesignThe actual staff/architect interview, made real: "Design an AI assistant over 10 years of our wiki + support tickets + code."

Start where you are.

Your progress is saved as you go, and every lesson you finish counts toward every path that includes it.

Compare paths