AI Automation Engineer for Lead Management and Client Operations

PROJECT OVERVIEW We are a professional services business looking for an AI automation engineer to improve two core areas of our operations: Lead intake, qualification, and follow-up Client request management and internal task coordination Our current processes involve manual data entry, inconsistent follow-up, information spread across multiple tools, and limited visibility into what requires attention. The goal is to build practical, reliable automations that reduce administrative work, improve response times, and help the business convert and serve more clients without adding unnecessary operational complexity. MILESTONE 1: LEAD INTAKE AND REVENUE FOLLOW-UP AUTOMATION Build a workflow that: • Captures inbound leads from the company’s existing forms, email, or other lead sources • Creates or updates lead records in the CRM • Removes duplicate records • Enriches and organizes lead information • Uses AI and business rules to classify or prioritize inquiries • Sends an immediate acknowledgment or creates a personalized response draft • Assigns the lead to the appropriate owner • Creates follow-up tasks and reminders • Supports meeting scheduling • Tracks lead status and follow-up completion The system should provide visibility into: • Number of new leads received • Median first-response time • Leads awaiting follow-up • Meetings scheduled • Lead stage and outcome MILESTONE 2: CLIENT OPERATIONS AND REQUEST TRIAGE Build a workflow that: • Captures client requests from the company’s existing communication channels • Classifies each request by client, topic, priority, and required action • Creates or updates the corresponding task or record • Assigns the request to the appropriate team member • Generates a suggested response when appropriate • Sends reminders for unresolved or overdue items • Maintains a clear status history • Produces a daily or weekly operations summary The system should provide visibility into: • Number and type of client requests • First-response time • Open and overdue items • Work assigned by team member • Requests completed • Common operational bottlenecks TECHNICAL EXPECTATIONS The final architecture may use: • n8n or Make • OpenAI API or Claude • CRM and email integrations • Google Workspace • APIs and webhooks • Supabase, PostgreSQL, Airtable, or the company’s existing database The exact tools should be selected based on the company’s current systems and operational needs. DELIVERABLES • Current-state workflow map • Proposed automation architecture • Two working automation workflows • CRM, email, calendar, and data integrations as required • Human approval steps for sensitive actions • Error handling, logging, and retry paths • Testing with realistic business scenarios • KPI baseline and post-implementation measurement • Technical documentation • Operator runbook • Recorded handoff walkthrough • Team training session ACCEPTANCE CRITERIA The project will be considered complete when: • Inbound leads are captured and tracked without repeated manual entry • Lead ownership and follow-up status are visible • Client requests are categorized and assigned consistently • Failed automations generate a visible error or alert • Sensitive external communications require approval unless explicitly authorized • The workflows are documented and can be operated by the internal team • Before-and-after performance metrics are recorded using actual business data PROJECT STRUCTURE Milestone 1: Discovery, workflow mapping, and KPI baseline Milestone 2: Lead intake and CRM follow-up automation Milestone 3: Client operations and request-triage automation Milestone 4: Testing, documentation, measurement, and handoff REQUIRED EXPERIENCE • AI workflow automation • n8n or Make • API and webhook integrations • CRM automation • OpenAI or Claude integration • Process mapping • Error handling and logging • Documentation and team handoff