Prepared for your DPA lead generation project — April 2026
Cold outreach about DPA to Texas teachers converts differently than outreach to firefighters — different qualifying criteria, different objections, different data sources. This system is built to handle that.
I build AI agent systems that run real business operations — lead research, outbound pipelines, CRM integrations. Not prototypes. Production systems that teams use daily. I specialise in Claude Code agent architectures: manager agents that coordinate specialist teams, each with custom-built tooling for your APIs and databases.
The system researches companies across multiple data sources, scores them against custom criteria, and surfaces qualified leads in a live dashboard. Same pattern, different domain.
These tools work in isolation. Every connection between agents has to be manually wired by a developer — and once wired, it's rigid. When a reply comes in, the n8n workflow doesn't know what the research agent found about that prospect. Sharing context between steps is brittle. Adding a new profession means duplicating entire workflows. It operates as a pipeline, but what you need is a team.
Claude Code agent teams communicate natively — no manual wiring, no rigid connections. The manager agent delegates to specialists who share context naturally. The research agent knows what the outreach agent sent because they're part of the same team, not separate workflows stitched together. Adding a new profession or city is a conversation with the manager, not a maintenance burden.
API-based systems charge per call — every research query, every classification, every LLM decision. A Claude subscription is flat, no matter how many campaigns you run.
Owns your entire outbound operation. Creates campaigns, delegates to specialists, monitors performance, learns what's working, and comes to you when it needs a decision.
Finds prospects by profession and location. Verifies emails before they enter the pipeline.
Generates personalised sequences tailored to each audience. Pushes campaigns through your email platform.
Classifies every reply — interested, objection, not interested, out of office. Flags uncertain ones for your review.
Routes qualified leads to your CRM and sends a Calendly booking link. Updates lead status across the pipeline.
You say "target Dallas nurses" — the manager contextualises the campaign for that specific audience.
The researcher finds matching prospects from your data providers and verifies every email address.
The outreach agent generates personalised sequences and pushes them through Smartlead.
Every reply is classified — interested, objection, not now. Qualified leads go straight to your CRM with a Calendly booking link.
The manager tracks what's working and adjusts targeting, messaging, and timing across campaigns.
Manager agent + specialist team, each with custom-built tooling for your APIs, database, and email platform. End-to-end from prospect research to Calendly booking. First campaign running within three weeks.
See the agents' work as it happens — campaign progress, lead status, reply classifications. Review results and give feedback that the manager agent acts on.
Full documentation as AI-readable markdown. The system can explain itself to you, improve over time, and create new campaigns on demand — no rebuilds, no rehiring.
Reply on Upwork and I'll send over a detailed scope within 24 hours.
Happy to jump on a call too — I'd love to discuss which profession to start with and your current Smartlead setup.