AI 모델 & 사용처
이 페이지는 한 질문에 답합니다: PulseAd 전반에서 AI는 실제로 어디에 쓰이고, 어디에 의도적으로 쓰이지 않는가? 모든 모델을 그 위에서 도는 에이전트에 매핑하고, 이를 뒤집어 모든 에이전트와 그 뒤의 모델을 보여주며, 제공자나 텍스트로 필터할 수 있습니다. 데이터는 코드로 검증됨.
한눈에
섹션 제목: “한눈에”PulseAd는 하나의 모델로 돌지 않습니다 — 서비스별로 의도적으로 분할됩니다:
ai-api(챗, 리포트, 플래닝) → OpenAI GPT‑5 패밀리(gpt-5,gpt-5.5,gpt-5.2,gpt-5-mini,gpt-5-nano,gpt-4.1/-mini/-nano), 검색어 세그먼테이션에 Google 모델 하나(gemini-2.5-flash-lite).pulsead-agents(자율 캠페인 운영) → AWS Bedrock의 Anthropic Claude — 기본 Haiku 4.5, 결정/작성 에이전트는 Sonnet 4.6.- Snowflake Cortex(시맨틱 데이터 레이어) → Claude Sonnet 4.6.
- 크리에이티브 → 이미지는 Google Gemini, 비디오는 멀티 벤더 게이트웨이(Sora 2, Runway, Kling, Google Veo, Grok).
패턴: 분류·수집에는 싸고 빠른 모델, 결정·작성에는 더 강한 모델. 라우터와 모니터는 가장 가벼운 모델로, 예산 결정·리포트 작성·SQL 생성은 가장 강한 모델로 동작합니다.
모델 → 에이전트와 에이전트 → 모델을 전환하고, 제공자로 필터하거나 검색하세요. 각 에이전트는 역할 배지를 가집니다: read(분석만), propose(승인용 변경 제안), write(실행, 게이트), internal(PulseAd 자체 상태만 기록), generate(크리에이티브 생성).
40AI 에이전트
19모델
7제공자
AI가 도는 곳
gpt-4.1 OpenAI 1 개 - Report streaming / metadata Reports read Report narrative text and metadata
gpt-4.1-mini OpenAI 2 개 - Action Router Chat read Classifies each message and resolves context
- Image / PDF summarizer Utility read Vision extraction of uploaded attachments
gpt-4.1-nano OpenAI 1 개 - Data summarizer Utility read Misc. structured data summaries
gpt-5 OpenAI 1 개 - Action sub-agent Chat write The only agent that changes an Amazon account (gated)
gpt-5-mini OpenAI 1 개 - Task sub-agent Chat internal Manages the internal inbox & tasks
gpt-5-nano OpenAI 1 개 - Memory classifier Chat internal Background — builds durable brand memory (16-card)
gpt-5.2 OpenAI 2 개 - PSM researcher / Pulson engine Chat read Brand-facing chat engine (Growth Twin)
- Dashboard Report Agent Reports read The report engine — fetch data, plan directives
gpt-5.5 OpenAI 5 개 - Query sub-agent Chat read Answers about your own performance from Snowflake
- Insight sub-agent Chat read External market / competitor / keyword lookups
- Action sub-agent Chat write The only agent that changes an Amazon account (gated)
- Composer Chat read Writes the final user-facing reply
- Performance Diagnose Reports read Finds declining campaigns, recommends actions
Sora 2 OpenAI 1 개 - video-generator (gateway) Creative generate One brief routed across video providers
claude-haiku-4-5 Anthropic 12 개 - monitoring (Campaign Health) Autonomous ops read Daily anomaly detection
- diagnoser (Campaign Health) Autonomous ops propose Root-cause; proposes structural actions
- rtb_analyst (Campaign Health) Autonomous ops propose RTB rule-group tuning proposals
- executor (Campaign Health) Autonomous ops write Applies approved bid/budget/pause actions
- oc_monitor (Optimize Cycle) Autonomous ops read Collects 7-day + MTD performance
- data_preparer (Weekly Report) Autonomous ops read Collects weekly performance data
- context_builder (Weekly Report) Autonomous ops read Aggregates operational context
- publisher (Weekly Report) Autonomous ops write Renders to HTML, publishes to S3 + Slack
- ads_scheduler Autonomous ops read Parses NL request into a scheduled task
- ads_executor Autonomous ops write Runs due scheduled changes (gated)
- amc_router Autonomous ops read Routes AMC requests to advisor/writer
- amc_advisor Autonomous ops read Answers conceptual AMC questions
claude-sonnet-4-6 Anthropic 9 개 - oc_input_checker (Optimize Cycle) Autonomous ops read Validates brand prerequisites
- oc_rebalancer (Optimize Cycle) Autonomous ops propose Computes & proposes product-group budgets
- oc_optimizer (Optimize Cycle) Autonomous ops propose Proposes campaign-level allocation
- oc_executor (Optimize Cycle) Autonomous ops write Applies budgets to Amazon (KISS-only live)
- oc_verifier (Optimize Cycle) Autonomous ops write Verifies & auto-corrects budget drift
- report_writer (Weekly Report) Autonomous ops read Writes the report draft
- amc_writer Autonomous ops read Generates & saves AMC SQL (no execute)
- keyword_harvester Autonomous ops propose Proposes profitable exact-match keywords
- waste_keyword_pauser Autonomous ops propose Proposes pausing zero-click spend
claude-sonnet-4-6 (Cortex) Anthropic 1 개 - AMAZON_ADS_AGENT (6 analysts) Semantic SQL read Text-to-SQL analytics over the data model
Gemini Google 2 개 - banana-bunch (image gen) Creative generate Product image + prompt → many images
- ads-variation (banner) Creative generate PSD → sized banner variations
gemini-2.5-flash-lite Google 1 개 - Search-term segmentation Utility read Classifies search terms + brand recognition
Google Veo Google 1 개 - video-generator (gateway) Creative generate One brief routed across video providers
Grok xAI 1 개 - video-generator (gateway) Creative generate One brief routed across video providers
Runway Runway 1 개 - video-generator (gateway) Creative generate One brief routed across video providers
Kling Kling 1 개 - video-generator (gateway) Creative generate One brief routed across video providers
Exa.ai (search) Exa.ai 1 개 - News & Market Intelligence Planning read News briefings (Exa.ai search + LLM summary)
Chat
- Action Router read
gpt-4.1-miniClassifies each message and resolves context - Query sub-agent read
gpt-5.5Answers about your own performance from Snowflake - Insight sub-agent read
gpt-5.5External market / competitor / keyword lookups - Action sub-agent write
gpt-5.5gpt-5The only agent that changes an Amazon account (gated) - Task sub-agent internal
gpt-5-miniManages the internal inbox & tasks - Composer read
gpt-5.5Writes the final user-facing reply - Memory classifier internal
gpt-5-nanoBackground — builds durable brand memory (16-card) - PSM researcher / Pulson engine read
gpt-5.2Brand-facing chat engine (Growth Twin)
Reports
- Dashboard Report Agent read
gpt-5.2The report engine — fetch data, plan directives - Report streaming / metadata read
gpt-4.1Report narrative text and metadata - Performance Diagnose read
gpt-5.5Finds declining campaigns, recommends actions
Planning
- News & Market Intelligence read
Exa.ai (search)News briefings (Exa.ai search + LLM summary)
Utility
- Search-term segmentation read
gemini-2.5-flash-liteClassifies search terms + brand recognition - Image / PDF summarizer read
gpt-4.1-miniVision extraction of uploaded attachments - Data summarizer read
gpt-4.1-nanoMisc. structured data summaries
Autonomous ops
- monitoring (Campaign Health) read
claude-haiku-4-5Daily anomaly detection - diagnoser (Campaign Health) propose
claude-haiku-4-5Root-cause; proposes structural actions - rtb_analyst (Campaign Health) propose
claude-haiku-4-5RTB rule-group tuning proposals - executor (Campaign Health) write
claude-haiku-4-5Applies approved bid/budget/pause actions - oc_input_checker (Optimize Cycle) read
claude-sonnet-4-6Validates brand prerequisites - oc_monitor (Optimize Cycle) read
claude-haiku-4-5Collects 7-day + MTD performance - oc_rebalancer (Optimize Cycle) propose
claude-sonnet-4-6Computes & proposes product-group budgets - oc_optimizer (Optimize Cycle) propose
claude-sonnet-4-6Proposes campaign-level allocation - oc_executor (Optimize Cycle) write
claude-sonnet-4-6Applies budgets to Amazon (KISS-only live) - oc_verifier (Optimize Cycle) write
claude-sonnet-4-6Verifies & auto-corrects budget drift - data_preparer (Weekly Report) read
claude-haiku-4-5Collects weekly performance data - context_builder (Weekly Report) read
claude-haiku-4-5Aggregates operational context - report_writer (Weekly Report) read
claude-sonnet-4-6Writes the report draft - publisher (Weekly Report) write
claude-haiku-4-5Renders to HTML, publishes to S3 + Slack - ads_scheduler read
claude-haiku-4-5Parses NL request into a scheduled task - ads_executor write
claude-haiku-4-5Runs due scheduled changes (gated) - amc_router read
claude-haiku-4-5Routes AMC requests to advisor/writer - amc_advisor read
claude-haiku-4-5Answers conceptual AMC questions - amc_writer read
claude-sonnet-4-6Generates & saves AMC SQL (no execute) - keyword_harvester propose
claude-sonnet-4-6Proposes profitable exact-match keywords - waste_keyword_pauser propose
claude-sonnet-4-6Proposes pausing zero-click spend
Semantic SQL
- AMAZON_ADS_AGENT (6 analysts) read
claude-sonnet-4-6 (Cortex)Text-to-SQL analytics over the data model
Creative
- banana-bunch (image gen) generate
GeminiProduct image + prompt → many images - ads-variation (banner) generate
GeminiPSD → sized banner variations - video-generator (gateway) generate
Sora 2RunwayKlingGoogle VeoGrokOne brief routed across video providers
No matches.
AI가 쓰이지 않는 곳 (의도적으로)
섹션 제목: “AI가 쓰이지 않는 곳 (의도적으로)”제품 리더에게 똑같이 중요한 점: 실행·안전 경로는 AI가 아니라 결정론적 코드입니다. 의도된 설계로, 동작을 예측 가능하고 감사 가능하게 만듭니다. 다음은 어떤 LLM도 결정하지 않습니다:
- 범위 & 인증 — 팀 + country 해석, 역할 체크,
404-존재-숨김 규칙. - 승인 & 적용 머신 — action-request 라이프사이클, 승인 게이트, Amazon으로의 디스패치, viewer 역할 차단. AI는 제안하고, 결정론적 코드가 적용합니다.
- 예산 계산 — 리밸런스 계산기는 모델이 아니라 순수 함수.
KISS-only 라이브 화이트리스트와 브랜드별amazon_ads_mode— 하드코딩 게이트.- 리포트 직렬화 — 최종 렌더 단계는 LLM 없이 고정 디렉티브를 냄(LLM은 내용만 계획).
- 메모리 기록 — 분류기는 AI지만, 메모리를 저장하는 dispatcher/effects는 결정론적.
- 데이터 플러밍 — Snowflake
SELECT/WITH-only 가드, off-Amazon 적재(Apps Script + SQL), 리포트 집계 엔드포인트. - 검색 ≠ 생성 — Exa.ai와 Tavily는 소스를 가져올 뿐, LLM은 요약만 함.
읽는 법
섹션 제목: “읽는 법”- AI는 추론과 콘텐츠에 집중: 요청 이해, 퍼포먼스 분석, 리포트 작성, 변경 제안, 크리에이티브 생성.
- AI는 실행에서 배제: 실제로 Amazon 계정을 바꾸거나, 돈을 움직이거나, 규칙을 강제하는 것은 사람 게이트 뒤의 결정론적 코드로 동작.
- 모델 선택이 의도를 신호: Haiku/
gpt-*-mini/-nano에이전트는 싼 분류기·수집기, Sonnet/gpt-5/gpt-5.2에이전트는 결정·작성 단계.
모델과 할당은 제품 발전에 따라 바뀝니다 — /sync-docs를 재실행해 소스에서 갱신하세요.