The AI Agency Pipeline

Conducts structured discovery interviews with clients. Produces AI Opportunity Maps that identify where AI can add value, with prioritised opportunities, expected impact, and a 90-day action plan.

Surfaces booking links at pipeline handoffs so contacts can schedule next steps without leaving the flow. Integrated with Cal.com and Google Calendar.

Transforms Interviewer output into a structured solution design — architecture decisions, tech stack recommendations, user stories, and component breakdowns.

The meta-agent that runs the pipeline itself. Knows what phase a project is in, what needs to happen next, and routes work between agents.

A five-agent pipeline that builds, tests, and critiques solutions. Generates → Tests → Critiques → Iterates without manual intervention between stages.

Designs, scaffolds, and deploys websites. Starting with this portfolio. Progresses to client-facing delivery with AI-powered pages.

Autonomous topic research agent. Given a topic, scope, and target media types, it searches the web, community discussions, and other sources — returning a structured report without human involvement during the run.

Transforms Knowledge Base content into audience-facing content. v1 produces LinkedIn post series from KB entries. v2 expands to multi-platform social. v3 targets YouTube scripts and training curricula.

Rapid domain onboarding from KB content or fresh ingestion. Given a niche or topic, produces a crash course package — concept brief, flashcard set, and self-quiz — so the pipeline is ready before client engagement.

Ingests podcast episodes into the knowledge base via Whisper transcription. Accepts episode URLs or RSS feeds and produces structured KB entries using the same pipeline as the YouTube ingestion adapter.

Surfaces maintenance tasks not already handled by existing tooling — skill health checks, KB freshness audits, CLAUDE.md drift detection, and folder structure validation. Building only when the gap proves real.