Develop AI Email Outreach Agent
Budget / Salary$250–750
TypeFreelance project
LocationRemote
Posted1 hour ago
The Goal: I am looking for an AI Automation Specialist to build a multi-agent system that automates my B2B lead generation and email drafting. We are an eventing/hire company based in South Africa.
For Phase 1, I need a "human-in-the-loop" system that scrapes targeted local business contacts, researches their recent activity, and drafts highly personalized outreach emails into a Google Sheet or Drafts folder for my review.The Workflow You Will Build:Sourcing & Scraping: Target specific industries in Gauteng, South Africa (Event coordinators, Caterers, Furniture rental companies, Marketing Professionals/PAs). You will need to implement a scraper or use APIs (Apollo, Custom Search, Tavily) to find these leads.Context Gathering: The agent must briefly scan the prospect’s website or public presence to extract one personalized detail (e.g., a recent event they hosted, core service).Drafting (LLM Integration): Use an LLM (Claude 3.5 Sonnet, GPT-4o, or Gemini Pro) to write a short, conversational B2B email. The prompt will introduce our company (20 years in business), mention the personalized detail, and ask if they need assistance with upcoming year-end functions.Output (Human-in-the-Loop): Do NOT send the emails. The final output must be logged into a Google Sheet (via API) or Gmail Drafts folder containing: Contact Name, Email, Company, Personalization Context, and the Drafted Email.
Preferred Tech Stack:Language: Python Framework: CrewAI, LangGraph, or AutoGen LLM: Open to your recommendation, but prefer Claude 3.5 Sonnet or GPT-4o for drafting.Automation: Open to using n8n or Make.com if it simplifies the Google Sheets/Gmail integration.Future Scope (Phase 2):If Phase 1 is successful, the contract will be extended to build out automated sending, follow-up sequences, and additional agents for competitor analysis on social media.When applying, please answer the following:Have you built an AI scraping and outreach agent before? If so, please link a case study, GitHub repo, or quick Loom video showing it working.What tool or API will you use to bypass scraping blockers for finding B2B leads?Which framework (CrewAI, LangGraph, or other) would you recommend for this specific flow and why?
For Phase 1, I need a "human-in-the-loop" system that scrapes targeted local business contacts, researches their recent activity, and drafts highly personalized outreach emails into a Google Sheet or Drafts folder for my review.The Workflow You Will Build:Sourcing & Scraping: Target specific industries in Gauteng, South Africa (Event coordinators, Caterers, Furniture rental companies, Marketing Professionals/PAs). You will need to implement a scraper or use APIs (Apollo, Custom Search, Tavily) to find these leads.Context Gathering: The agent must briefly scan the prospect’s website or public presence to extract one personalized detail (e.g., a recent event they hosted, core service).Drafting (LLM Integration): Use an LLM (Claude 3.5 Sonnet, GPT-4o, or Gemini Pro) to write a short, conversational B2B email. The prompt will introduce our company (20 years in business), mention the personalized detail, and ask if they need assistance with upcoming year-end functions.Output (Human-in-the-Loop): Do NOT send the emails. The final output must be logged into a Google Sheet (via API) or Gmail Drafts folder containing: Contact Name, Email, Company, Personalization Context, and the Drafted Email.
Preferred Tech Stack:Language: Python Framework: CrewAI, LangGraph, or AutoGen LLM: Open to your recommendation, but prefer Claude 3.5 Sonnet or GPT-4o for drafting.Automation: Open to using n8n or Make.com if it simplifies the Google Sheets/Gmail integration.Future Scope (Phase 2):If Phase 1 is successful, the contract will be extended to build out automated sending, follow-up sequences, and additional agents for competitor analysis on social media.When applying, please answer the following:Have you built an AI scraping and outreach agent before? If so, please link a case study, GitHub repo, or quick Loom video showing it working.What tool or API will you use to bypass scraping blockers for finding B2B leads?Which framework (CrewAI, LangGraph, or other) would you recommend for this specific flow and why?
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.