Corpus-Complete Building Blocks
Snapshot: corpus-complete-20260711 (2026-07-11) · source commit 1737631198b184e05ba0a343d67df6fa9e71b938 Scope: 1780 approved schema-v2 extractions; 1754 deduped sources; 991 deduped agentic projects.
Corpus-complete analysis output. Counts are generated from extraction.json; citations link to deterministic report.md views. This is an input to, not the output of, Phase 4 synthesis.
A candidate qualifies when it appears in at least 5% of the deduped agentic corpus and spans at least 4 industries. Counts normalize only reviewer-confirmed aliases.
Agentic denominator: 991 deduped projects. Frequency threshold: 50 deduped projects.
Candidate Fundamental Building Blocks
| Pattern | Original names counted | Deduped projects | Frequency | Industries | Year trend | Evidence strength | Evidence provenance | Project provenance | Component signatures | Representative projects |
|---|---|---|---|---|---|---|---|---|---|---|
tool-use-orchestration |
tool-use-orchestration |
643 | 64.9% | 16 | 2023: 58 2024: 132 2025: 235 2026: 191 |
953 stated / 7 implied | 366 article / 478 transcript / 116 summary | 609 source-backed / 34 summary-only | application component (1362) evaluation/observability layer (851) context/memory layer (711) retrieval/search layer (687) |
2x-engineering-throughput-through-ai-first-development-platform (intercom, 2026) actionable-ci-intelligent-analysis-and-auto-remediation-of-ci-pipeline-failures (block, 2026) adopting-model-context-protocol-mcp-in-financial-services-for-ai-system-integration (evergreen_wealth_/bloomberg/_saxo_bank, 2026) |
eval-driven-development |
eval-driven-developmentfeedback-loop-iteration |
475 | 47.9% | 16 | 2023: 55 2024: 93 2025: 160 2026: 145 |
775 stated / 5 implied | 291 article / 404 transcript / 85 summary | 452 source-backed / 23 summary-only | application component (1043) evaluation/observability layer (1013) retrieval/search layer (526) context/memory layer (521) |
accelerating-llm-inference-with-speculative-decoding-for-ai-agent-applications (linkedin, 2025) agent-based-ai-assistants-for-enterprise-and-e-commerce-applications (prosus, 2024) agent-driven-ui-framework-migration-at-enterprise-scale (block, 2026) |
human-in-the-loop-gate |
human-in-the-loophuman-in-the-loop-gate |
396 | 40.0% | 16 | 2023: 45 2024: 76 2025: 138 2026: 113 |
597 stated / 5 implied | 207 article / 338 transcript / 57 summary | 383 source-backed / 13 summary-only | application component (988) evaluation/observability layer (646) context/memory layer (432) data/source layer (379) |
advanced-agent-monitoring-and-debugging-with-langsmith-integration (replit, 2024) agentic-ai-for-cloud-migration-and-application-modernization-at-scale (commonwealth_bank_of_australia, 2025) agentic-ai-system-for-construction-industry-tender-management-and-quote-generation (tendos_ai, 2026) |
multi-agent-orchestration |
multi-agent-orchestration |
332 | 33.5% | 16 | 2023: 30 2024: 58 2025: 117 2026: 111 |
479 stated / 6 implied | 159 article / 255 transcript / 71 summary | 318 source-backed / 14 summary-only | application component (792) evaluation/observability layer (468) retrieval/search layer (361) context/memory layer (347) |
agent-based-ai-assistants-for-enterprise-and-e-commerce-applications (prosus, 2024) agent-first-ai-development-platform-with-multi-surface-orchestration (google_deepmind, 2025) agentic-ai-architecture-for-investment-management-platform (blackrock, year n/a) |
observability-tracing |
observability-tracing |
329 | 33.2% | 16 | 2023: 29 2024: 61 2025: 122 2026: 98 |
538 stated / 3 implied | 182 article / 305 transcript / 54 summary | 314 source-backed / 15 summary-only | evaluation/observability layer (721) application component (679) model/inference layer (356) context/memory layer (354) |
2x-engineering-throughput-through-ai-first-development-platform (intercom, 2026) advanced-agent-monitoring-and-debugging-with-langsmith-integration (replit, 2024) ai-agent-driven-software-development-platform-for-enterprise-engineering-teams (factory, year n/a) |
rag-retrieval |
rag-retrieval |
304 | 30.7% | 15 | 2023: 46 2024: 89 2025: 108 2026: 41 |
436 stated / 3 implied | 190 article / 193 transcript / 56 summary | 289 source-backed / 15 summary-only | retrieval/search layer (638) application component (594) data/source layer (382) evaluation/observability layer (340) |
agentic-ai-for-title-operations-workflow-optimization (rocket, 2026) agentic-ai-manufacturing-reasoner-for-automated-root-cause-analysis (apollo_tyres, 2025) agentic-ai-system-for-document-summarization-and-analysis (moveworks, 2024) |
safety-policy-enforcement |
guardrailssafety-policy-enforcement |
211 | 21.3% | 16 | 2023: 17 2024: 42 2025: 78 2026: 65 |
358 stated / 1 implied | 141 article / 183 transcript / 35 summary | 207 source-backed / 4 summary-only | application component (458) evaluation/observability layer (311) context/memory layer (241) data/source layer (231) |
adopting-model-context-protocol-mcp-in-financial-services-for-ai-system-integration (evergreen_wealth_/bloomberg/_saxo_bank, 2026) advanced-fine-tuning-techniques-for-multi-agent-orchestration-at-scale (amazon, 2026) agent-identity-and-access-management-for-production-ai-systems (uber, 2026) |
agent-memory |
agent-memory |
183 | 18.5% | 15 | 2023: 13 2024: 33 2025: 57 2026: 75 |
279 stated / 1 implied | 115 article / 141 transcript / 24 summary | 173 source-backed / 10 summary-only | application component (390) context/memory layer (327) evaluation/observability layer (256) retrieval/search layer (224) |
agent-memory-system-for-personalized-food-ordering-and-discovery (doordash, 2026) agentic-ai-for-legal-research-building-deep-research-in-westlaw-and-cocounsel (thomson_reuters, 2025) agentic-ai-systems-for-legal-tax-and-compliance-workflows (thomson_reuters, 2024) |
model-routing |
model-routing |
169 | 17.1% | 14 | 2023: 23 2024: 38 2025: 52 2026: 45 |
265 stated / 2 implied | 89 article / 148 transcript / 30 summary | 164 source-backed / 5 summary-only | application component (333) model/inference layer (284) evaluation/observability layer (279) context/memory layer (192) |
agent-based-workflow-automation-in-spreadsheets-for-non-technical-users (otto, 2025) ai-driven-media-analysis-and-content-assembly-platform-for-large-scale-video-archives (bloomberg_media, 2024) ai-employee-agent-operating-in-slack-with-multi-tool-integration (viktor, 2026) |
structured-output-contract |
structured-output-contract |
158 | 15.9% | 14 | 2023: 23 2024: 33 2025: 59 2026: 35 |
249 stated / 2 implied | 116 article / 112 transcript / 23 summary | 153 source-backed / 5 summary-only | application component (332) evaluation/observability layer (217) data/source layer (213) context/memory layer (178) |
ai-agent-system-for-automated-security-investigation-and-alert-triage (slack, 2025) ai-native-multi-agent-system-for-customer-onboarding-and-kyc (brex, 2026) ai-orchestrated-code-review-system-at-scale (cloudflare, 2026) |
llm-as-judge |
llm-as-judge |
120 | 12.1% | 13 | 2023: 16 2024: 21 2025: 43 2026: 31 |
179 stated / 1 implied | 69 article / 93 transcript / 18 summary | 115 source-backed / 5 summary-only | evaluation/observability layer (346) application component (228) retrieval/search layer (182) context/memory layer (135) |
agent-testing-and-evaluation-using-autonomous-vehicle-simulation-principles (coval, 2023) agentic-ai-search-with-custom-evaluation-framework-for-church-management (pushpay, 2026) ai-driven-code-review-agent-reduces-pr-cycle-time-by-308 (atlassian, 2026) |
cost-aware-routing |
cost-aware-routing |
112 | 11.3% | 14 | 2023: 15 2024: 19 2025: 36 2026: 37 |
179 stated / 0 implied | 63 article / 89 transcript / 27 summary | 106 source-backed / 6 summary-only | application component (224) model/inference layer (200) evaluation/observability layer (195) retrieval/search layer (158) |
ai-agent-for-automated-merchant-classification-correction (ramp, 2025) ai-agent-system-for-automated-b2b-research-and-sales-pipeline-generation (unify, 2025) ai-agents-for-ml-experiment-orchestration-reducing-friction-in-machine-learning-workflows (teads, 2026) |
latency-optimization |
latency-optimization |
106 | 10.7% | 10 | 2023: 16 2024: 18 2025: 44 2026: 24 |
179 stated / 0 implied | 54 article / 114 transcript / 11 summary | 101 source-backed / 5 summary-only | application component (240) model/inference layer (159) retrieval/search layer (144) context/memory layer (134) |
accelerating-llm-inference-with-speculative-decoding-for-ai-agent-applications (linkedin, 2025) agentic-ai-architecture-for-meeting-intelligence-and-productivity-automation (zoom, 2025) agentic-news-analysis-platform-for-digital-asset-market-making (fsi, 2025) |
guardrails-io-filtering |
guardrails-io-filtering |
99 | 10.0% | 14 | 2023: 10 2024: 22 2025: 45 2026: 17 |
158 stated / 1 implied | 62 article / 89 transcript / 8 summary | 95 source-backed / 4 summary-only | application component (195) evaluation/observability layer (145) retrieval/search layer (122) context/memory layer (118) |
ai-agent-for-automated-merchant-classification-and-transaction-matching (ramp, 2025) ai-augmented-code-review-system-for-large-scale-software-development (uber, 2025) ai-powered-code-review-platform-at-scale (uber, 2025) |
context-compaction |
context-compaction |
98 | 9.9% | 10 | 2023: 5 2024: 9 2025: 39 2026: 42 |
156 stated / 0 implied | 61 article / 83 transcript / 12 summary | 92 source-backed / 6 summary-only | application component (216) context/memory layer (198) evaluation/observability layer (153) retrieval/search layer (113) |
agentic-code-reviewers-as-system-protectors (block, 2026) ai-agents-for-documenting-tribal-knowledge-in-large-scale-data-pipelines (meta, 2026) ai-powered-code-review-platform-using-abstract-syntax-trees-and-llm-context (baz, 2023) |
data-quality-loop |
data-quality-loop |
98 | 9.9% | 12 | 2023: 9 2024: 25 2025: 32 2026: 23 |
161 stated / 3 implied | 49 article / 107 transcript / 8 summary | 94 source-backed / 4 summary-only | application component (234) evaluation/observability layer (210) retrieval/search layer (138) data/source layer (129) |
ai-agent-automation-of-security-operations-center-analysis (doppel, 2025) ai-agents-for-data-labeling-and-infrastructure-maintenance-at-scale (plaid, 2025) ai-powered-artwork-quality-moderation-and-streaming-quality-management-at-scale (amazon_prime_video, 2025) |
durable-execution |
durable-execution |
93 | 9.4% | 8 | 2023: 5 2024: 11 2025: 26 2026: 47 |
151 stated / 1 implied | 49 article / 92 transcript / 11 summary | 90 source-backed / 3 summary-only | application component (207) evaluation/observability layer (126) context/memory layer (117) workflow/messaging layer (98) |
ai-agent-for-automated-feature-flag-removal (duolingo, 2025) ai-driven-ddos-protection-system-using-temporal-workflow-orchestration (salesforce, 2026) ai-generated-trip-reports-for-outdoor-recreation-guides (guidesly, 2026) |
fine-tuning-adaptation |
fine-tuning-adaptation |
90 | 9.1% | 12 | 2023: 19 2024: 21 2025: 22 2026: 19 |
132 stated / 1 implied | 31 article / 88 transcript / 14 summary | 86 source-backed / 4 summary-only | model/inference layer (194) application component (171) evaluation/observability layer (138) retrieval/search layer (117) |
advanced-fine-tuning-techniques-for-multi-agent-orchestration-at-scale (amazon, 2026) agentic-news-analysis-platform-for-digital-asset-market-making (fsi, 2025) ai-agent-automation-of-security-operations-center-analysis (doppel, 2025) |
privacy-preserving-processing |
privacy-preserving-processing |
88 | 8.9% | 13 | 2023: 9 2024: 14 2025: 36 2026: 24 |
135 stated / 0 implied | 54 article / 79 transcript / 2 summary | 87 source-backed / 1 summary-only | application component (188) data/source layer (133) context/memory layer (103) model/inference layer (103) |
ai-agent-system-for-automated-design-system-documentation (uber, 2026) ai-powered-accounting-automation-using-claude-and-amazon-bedrock (floqast, 2025) ai-powered-call-center-agents-for-healthcare-operations (heyrevia, 2023) |
context-engineering |
context-engineering |
83 | 8.4% | 13 | 2023: 6 2024: 11 2025: 33 2026: 30 |
124 stated / 1 implied | 40 article / 49 transcript / 36 summary | 77 source-backed / 6 summary-only | application component (177) evaluation/observability layer (121) context/memory layer (115) retrieval/search layer (74) |
building-claude-code-scaling-ai-powered-development-from-terminal-prototype-to-production (anthropic, 2024) harness-engineering-building-software-where-humans-steer-and-agents-execute (openai, 2025) harness-engineering-for-agentic-coding-systems (langchain, 2026) |
fallback-degradation |
fallback-degradation |
60 | 6.1% | 9 | 2023: 9 2024: 11 2025: 24 2026: 13 |
94 stated / 1 implied | 45 article / 42 transcript / 8 summary | 57 source-backed / 3 summary-only | application component (174) model/inference layer (83) context/memory layer (81) evaluation/observability layer (69) |
actionable-ci-intelligent-analysis-and-auto-remediation-of-ci-pipeline-failures (block, 2026) ai-powered-contact-center-transformation-with-amazon-connect (traeger, 2024) ai-powered-help-desk-for-accounts-payable-automation (xelix, 2025) |
prompt-versioning |
prompt-versioning |
59 | 6.0% | 9 | 2023: 10 2024: 11 2025: 25 2026: 12 |
89 stated / 2 implied | 44 article / 35 transcript / 12 summary | 55 source-backed / 4 summary-only | evaluation/observability layer (133) context/memory layer (123) application component (99) model/inference layer (78) |
agent-driven-development-for-ai-research-using-github-copilot-cli (github, 2026) ai-agent-optimization-using-claude-to-systematically-improve-memory-extraction-quality (lerim, 2026) ai-powered-hyper-personalized-email-campaign-automation (promptlayer, 2025) |
prompt-chaining |
prompt-chaining |
52 | 5.2% | 7 | 2023: 10 2024: 15 2025: 16 2026: 9 |
82 stated / 0 implied | 49 article / 26 transcript / 7 summary | 50 source-backed / 2 summary-only | application component (111) context/memory layer (88) evaluation/observability layer (63) data/source layer (55) |
agentic-ai-search-with-custom-evaluation-framework-for-church-management (pushpay, 2026) agentic-ai-system-for-document-summarization-and-analysis (moveworks, 2024) ai-augmented-code-review-system-for-large-scale-software-development (uber, 2025) |
Near-Threshold Patterns
These patterns reach at least 3.5% frequency or clear the frequency threshold while missing industry spread.
| Pattern | Original names counted | Deduped projects | Frequency | Industries | Missing criterion | Representative projects |
|---|---|---|---|---|---|---|
retrieval-ranking |
retrieval-ranking |
49 | 4.9% | 9 | frequency | agent-memory-system-for-personalized-food-ordering-and-discovery (doordash, 2026) ai-powered-conversational-food-ordering-with-ilo (ifood, 2026) |
batch-offline-processing |
batch-offline-processing |
41 | 4.1% | 10 | frequency | agentic-mapreduce-for-whole-codebase-security-scanning (devin, 2026) ai-powered-automated-graphql-schema-cleanup (whatnot, 2025) |
confidence-thresholding |
confidence-thresholding |
39 | 3.9% | 6 | frequency | ai-driven-digital-twins-for-industrial-infrastructure-optimization (geminus, 2025) ai-native-multi-agent-system-for-customer-onboarding-and-kyc (brex, 2026) |
hybrid-search |
hybrid-search |
39 | 3.9% | 12 | frequency | agentic-ai-system-for-document-summarization-and-analysis (moveworks, 2024) agentic-rag-implementation-for-retail-personalization-and-customer-support (mongodb, 2024) |
Other Non-Qualifying Patterns
The highest-support remaining names are shown for boundary review.
| Pattern | Deduped projects | Frequency | Industries | Missing criterion |
|---|---|---|---|---|
query-rewriting |
31 | 3.1% | 10 | frequency |
semantic-caching |
31 | 3.1% | 6 | frequency |
synthetic-data-generation |
31 | 3.1% | 8 | frequency |
canary-rollout |
27 | 2.7% | 6 | frequency |
async-message-orchestration |
22 | 2.2% | 5 | frequency |
workflow-decomposition |
14 | 1.4% | 4 | frequency |
model-distillation |
11 | 1.1% | 5 | frequency |
forward-deployed-engineering |
3 | 0.3% | 1 | frequency, industry spread |
structured-output-validation |
2 | 0.2% | 1 | frequency, industry spread |
code-execution-sandbox |
1 | 0.1% | 1 | frequency, industry spread |
fallback-strategy |
1 | 0.1% | 1 | frequency, industry spread |
incremental-rollout |
1 | 0.1% | 1 | frequency, industry spread |
Reviewer Checks
- Confirm the 5% and 4-industry boundaries before pattern-page synthesis.
- Audit component signatures: they are derived from component role text and are intentionally coarse.
- Proposed vocabulary terms remain provisional until reviewer policy accepts them.