Start Here5 min readchapter
AI Software Factory Topic Index
Use this map when you have a topic in mind but do not know where it belongs in the curriculum. “Start here” gives the shortest useful orientation. “Deep dive” supplies the production architecture. “Practice” points to a lab or exercise wher
Risk: variableLifecycle: intent · plan · execute · verify · deliver · learnContent reviewed 2026-08-30Maturity guide →
Claim boundaryThis is curriculum guidance. It does not by itself prove a production implementation.
Use this map when you have a topic in mind but do not know where it belongs in the curriculum. “Start here” gives the shortest useful orientation. “Deep dive” supplies the production architecture. “Practice” points to a lab or exercise where one exists.
Factory and operating model
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| AI Software Factory | Factory overview | Vision | Chapter whiteboard |
| Agent Factory | Stack boundaries | Factory configuration | Capability-map lab |
| Mission Control | Lifecycle | Capability and admission map | Golden path |
| Human-agent operating model | Quick Read | Full operating model | Decision-packet exercise |
| Human-in/on/out-of-the-loop | Glossary | Compounding engineering | Attention-budget lab |
| Progressive autonomy | Trust calibration | Maturity model | Promotion/demotion exercise |
| Compounding engineering | Glossary | Compounding and attention | Correction-clustering lab |
| Capability coverage and maturity | Coverage map | Reviewer guide | External review checklist |
| Detailed architecture ownership | Coverage matrix | Reference architecture | Four-view trace exercise |
| System inventory and lifecycle | Inventory reference | Governance operating model | Inventory lifecycle exercise |
| Architecture and autonomy selection | Selection ladder | Operational autonomy | Simpler-alternative review |
Agent Factory and capability supply chain
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| Capability lifecycle and registries | Capability supply chain | Certification and retirement | Certification lab |
| Agent, skill, tool, and prompt versioning | Packaging and versioning | Execution manifests | Dependency-lock exercise |
| Capability discovery and resolution | Registries | Packaging and dependencies | Revocation exercise |
| Capability evaluation and promotion | Certification lifecycle | Evaluation science | Continual-improvement lab |
| Tool, skill, and integration contracts | Contract reference | Capability packaging | External capability lab |
Agents, harnesses, and interoperability
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| AI coding agents | Stack boundaries | Agent architecture | Agent-composition exercise |
| Inner and outer harnesses | Stack boundaries | Harnesses and adapters | Two-adapter comparison |
| Agent orchestration | Agent and loop patterns | Runtime orchestration | Router/verifier lab |
| Orchestration component contracts | Reference architecture | Runtime contracts | Failure and cost lab |
| Multi-agent topology and delegation | Topology reference | Architecture selection | Independence comparison |
| Generate–verify–retry–escalate | Loop engineering glossary | Agent and loop patterns | Oscillation exercise |
| MCP | Agent architecture | Protocols | Tool-gateway threat exercise |
| ACP, AG-UI, and A2A | Protocols table | Research canon | Protocol-boundary whiteboard |
| Claude Code and Codex | Dated case studies | Official references in the chapter | Adapter conformance lab |
Data, context, evaluation, and quality
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| Data Understanding | Glossary | Data and knowledge engineering | Source-quality lab |
| Knowledge Engineering and RAG | Agent context | Retrieval lifecycle | Hybrid-retrieval lab |
| Knowledge pipeline, poisoning, and revocation | Knowledge engineering | Pipeline specification | Knowledge revocation lab |
| Context Engineering | Quick Read | Context compilation | Context-package lab |
| Semantic Engineering | Glossary | Semantic contracts | Entity-resolution exercise |
| AI evaluations | Model routing | Evaluation engineering | Baseline/candidate lab |
| Trace replay and run comparison | Glossary | Replay and comparison | Trajectory-diff lab |
| Independent verification | Quality Quick Read | Quality contracts | Proof-package exercise |
| Evaluation datasets, calibration, and uncertainty | Evaluation science | Evaluation engineering | Repeated-trial lab |
| Prompt, skill, tool, and context optimization | Capability learning | Governed learning | Promotion lab |
Repository onboarding and autonomous workflows
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| Repository onboarding and readiness | Repository onboarding | Data and context | Onboarding lab |
| Codebase indexing and intelligence | Repository onboarding | Multi-repository delivery | Drift and refresh exercise |
| Autonomous workflow catalog and backlog | Workflow catalog | Factory economics | Three-workflow manifest lab |
| Feature, defect, test, dependency, and modernization work | Change workflows | Testing strategy | Reproduction and fault-sensitivity exercise |
| Security, incident, production, and documentation work | Operational workflows | Production verification | Incident lab |
Runtime, infrastructure, and delivery
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| Control plane | Quick Read | Runtime state machines | Authority-boundary whiteboard |
| Development environments | Stack boundaries | Environment and compute | Clean-environment lab |
| Compute infrastructure | Glossary | Worker-fleet design | Capacity/failover exercise |
| Composable, enterprise, and open-source infrastructure | Glossary | Build, buy, and bring your own | Vendor-boundary review |
| Sandboxing and isolation | Runtime map | Sandboxed execution | Escape and teardown tests |
| Reliability and recovery | Attempts and leases | Factory SRE | Failure-injection lab |
| Multi-repository development | Glossary | Coordinated delivery | Two-repository lab |
| Git submodules and subtrees | Strategy comparison | Git references in the chapter | Composition tradeoff exercise |
| Production feedback and reproduction | Factory SRE | Feedback-to-merge | Reproduction lab |
| Automated PR review and CodeRabbit | Feedback-to-merge | Dated product references in the chapter | Bounded review-loop exercise |
| Merge queues and agentic merge maintenance | Glossary | Merge maintenance | Currentness/conflict simulation |
| Unit, integration, contract, property, mutation, fuzz, performance, accessibility, and visual testing | Testing strategy | Quality contracts | Fault-sensitivity lab |
| CI/CD, artifacts, database migrations, and API compatibility | Delivery engineering | Supply-chain security | Migration-compatibility lab |
| Progressive delivery, feature flags, canaries, and rollback | Progressive delivery | Factory SRE | Rollback lab |
Platform, operator experience, and security
| Topic | Start here | Deep dive | Practice |
|---|---|---|---|
| Developer portal, service catalog, and golden paths | Platform product | Enterprise adoption | Golden-path usability exercise |
| Scheduling, quotas, capacity, fairness, and cost | Scheduling and cost | Runtime orchestration | Queue simulation |
| Resilience, disaster recovery, and emergency authority | Resilience and DR | Attempts and recovery | Disaster-recovery lab |
| Human-agent control surfaces and accessibility | Operator experience | Human attention | Control-state usability lab |
| Workflow DSLs, events, schema evolution, and storage | Workflow and event contracts | Runtime orchestration | Event replay and compensation exercise |
| Observability semantics, token cost, cardinality, and forensics | Observability semantics | Factory observability | Forensic-bundle exercise |
| Enterprise operations, reliability, and FinOps | Operations reference | Scheduling and cost | Failure and cost lab |
| Monitoring, drift, triage, and response | Control tower response | Observability semantics | Detection-to-closure exercise |
| Agentic threats, prompt injection, tool abuse, and memory poisoning | Agentic threat model | Security architecture | Attack and containment lab |
| Workload identity, secrets, privacy, licensing, and compliance | Identity and privacy | Governance and policy | Identity-chain exercise |
| Authority, emergency control, and recertification | Governance controls | Authority and emergency control | Containment and replay lab |
Choosing the next document
If you are still unsure, use the Reading Paths. If a term is unfamiliar, use the Canonical Glossary. Use Capability Coverage and Maturity to distinguish documented architecture from validated and operationally proven capability. If a claim concerns current product behavior, use the versioned Mission Control capability and admission map.
External review
Review this chapter.
Challenge a claim, boundary, missing failure mode, unclear term, or unsupported evidence statement.
- Claim
- Boundary
- Failure
- Evidence