Agent loop that critiques and revises its own output for marketplace moderation. Full trace capture via Langfuse, retry budget, and ship criteria.
Agent loop that critiques and revises its own output for code PR review. Full trace capture via Weights & Biases Weave, retry budget, and ship criteria.
Agent loop that critiques and revises its own output for customer support triage. Full trace capture via Helicone, retry budget, and ship criteria.
End-to-end agent debate agent implemented in Inngest agent-kit for bug triage from Sentry logs. Includes graph/state design, tool wiring, loop termination, observability via Langfuse, and evals.
End-to-end agent debate agent implemented in LangGraph for investor update drafting. Includes graph/state design, tool wiring, loop termination, observability via custom SQLite trace store, and evals.
End-to-end agent debate agent implemented in CrewAI for invoice reconciliation. Includes graph/state design, tool wiring, loop termination, observability via OpenTelemetry + Honeycomb, and evals.
End-to-end agent debate agent implemented in OpenAI Agents SDK for sales lead enrichment. Includes graph/state design, tool wiring, loop termination, observability via Braintrust, and evals.
End-to-end agent debate agent implemented in Claude Agent SDK for social media scheduling. Includes graph/state design, tool wiring, loop termination, observability via Langfuse, and evals.
End-to-end agent debate agent implemented in Mastra for recruiting resume screen. Includes graph/state design, tool wiring, loop termination, observability via custom SQLite trace store, and evals.
End-to-end agent debate agent implemented in Vercel AI SDK for competitive analysis. Includes graph/state design, tool wiring, loop termination, observability via OpenTelemetry + Honeycomb, and evals.
End-to-end agent debate agent implemented in Smolagents for daily research briefing. Includes graph/state design, tool wiring, loop termination, observability via Braintrust, and evals.
End-to-end agent debate agent implemented in LlamaIndex agents for SEO keyword research. Includes graph/state design, tool wiring, loop termination, observability via Langfuse, and evals.