AI models & usage

This page answers one question: where is AI actually used across PulseAd, and where is it deliberately not? It maps every model to the agents that run on it, flips that to every agent and the model behind it, and is filterable by provider or text. The data is code-verified.

At a glance

PulseAd doesn’t run on one model — it runs on a deliberate split by service:

• ai-api (chat, reports, planning) → OpenAI GPT‑5 family (gpt-5, gpt-5.5, gpt-5.2, gpt-5-mini, gpt-5-nano, gpt-4.1/-mini/-nano), with one Google model (gemini-2.5-flash-lite) for search-term segmentation.
• pulsead-agents (autonomous campaign ops) → Anthropic Claude on AWS Bedrock — Haiku 4.5 by default, Sonnet 4.6 for the decision/writer agents.
• Snowflake Cortex (the semantic data layer) → Claude Sonnet 4.6.
• Creative → Google Gemini for images, and a multi-vendor video gateway (Sora 2, Runway, Kling, Google Veo, Grok).

The pattern: cheap, fast models for classification and collection; stronger models for decisions and writing. Router and monitors run on the lightest models; budget decisions, report writing, and SQL generation run on the strongest.

Explore it

Toggle between by model → agents and by agent → models, filter by provider, or search. Each agent carries a role badge: read (analyzes only), propose (suggests a change for approval), write (executes, gated), internal (writes only to PulseAd’s own state), and generate (produces creative).

40AI agents

19models

7providers

Where the AI runs

Anthropic 22 OpenAI 14 Google 4 xAI 1 Runway 1 Kling 1 Exa.ai 1

By model By agent

All OpenAIAnthropicGooglexAIRunwayKlingExa.ai

gpt-4.1 OpenAI 1 agents

• Report streaming / metadata Reports read Report narrative text and metadata

gpt-4.1-mini OpenAI 2 agents

• 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 agents

• Data summarizer Utility read Misc. structured data summaries

gpt-5 OpenAI 1 agents

• Action sub-agent Chat write The only agent that changes an Amazon account (gated)

gpt-5-mini OpenAI 1 agents

• Task sub-agent Chat internal Manages the internal inbox & tasks

gpt-5-nano OpenAI 1 agents

• Memory classifier Chat internal Background — builds durable brand memory (16-card)

gpt-5.2 OpenAI 2 agents

• 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 agents

• 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 agents

• video-generator (gateway) Creative generate One brief routed across video providers

claude-haiku-4-5 Anthropic 12 agents

• 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 agents

• 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 agents

• AMAZON_ADS_AGENT (6 analysts) Semantic SQL read Text-to-SQL analytics over the data model

Gemini Google 2 agents

• 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 agents

• Search-term segmentation Utility read Classifies search terms + brand recognition

Google Veo Google 1 agents

• video-generator (gateway) Creative generate One brief routed across video providers

Grok xAI 1 agents

• video-generator (gateway) Creative generate One brief routed across video providers

Runway Runway 1 agents

• video-generator (gateway) Creative generate One brief routed across video providers

Kling Kling 1 agents

• video-generator (gateway) Creative generate One brief routed across video providers

Exa.ai (search) Exa.ai 1 agents

• News & Market Intelligence Planning read News briefings (Exa.ai search + LLM summary)

Where AI is not used (by design)

Just as important for a product leader: the execution and safety paths are deterministic, not AI. That’s intentional — it makes behavior predictable and auditable. No LLM decides these:

• Scope & auth — resolving team + country, role checks, the 404-hides-existence rule.
• The approval & apply machinery — the action-request lifecycle, the approval gates, the dispatch to Amazon, the viewer-role block. AI proposes; deterministic code applies.
• Budget math — the rebalance calculator is a pure function, not a model.
• The KISS-only live whitelist and per-brand amazon_ads_mode — hard-coded gates.
• Report serialization — the report’s final render step emits fixed directives with no LLM (the LLM only plans the content).
• Memory write-back — the classifier is AI, but the dispatcher/effects that persist memory are deterministic.
• Data plumbing — Snowflake’s SELECT/WITH-only guard, off-Amazon ingestion (Apps Script + SQL), the report aggregator endpoints.
• Retrieval ≠ generation — Exa.ai and Tavily fetch sources; the LLM only summarizes them.

How to read this

• AI is concentrated in reasoning and content: understanding a request, analyzing performance, writing reports, proposing changes, generating creative.
• AI is kept out of execution: anything that actually changes an Amazon account, moves money, or enforces a rule runs on deterministic code behind a human gate.
• The model choice signals intent: a Haiku/gpt-*-mini/-nano agent is a cheap classifier or collector; a Sonnet/gpt-5/gpt-5.2 agent is a decision or writing step.

Models and assignments change as the product evolves — re-run /sync-docs to refresh this from source.