AI-powered WhatsApp service, qualification and follow-up — 24/7

Problem

Leads arriving on WhatsApp at any hour with no qualification, no CRM record and no consistent follow-up — pipeline silently evaporating.

Solution

Multimodal n8n agent (text, audio via Whisper, image via Vision) with a Redis message buffer, Postgres chat memory and Supabase lead registry. Qualifies consultatively, moves Kommo stages and tags, uses dedicated tools (qualification, name update, human handoff) and emits funnel events to the dashboard. Cascading follow-ups (10 min → 14 days) auto-close silent conversations and resume them when the lead replies.

Result

Live across multiple client operations; 24/7 service with AI→human handoff and an end-to-end tracked funnel.

Scheduling and knowledge agent for a salon

Problem

Salon needed 24/7 service and scheduling without blowing up API cost and latency.

Solution

Hybrid RAG: static info served from a vector store, live API reserved only for scheduling. Daily embedding ingestion cron and AI→human handoff. Reverse-engineered an undocumented third-party API (Cognito auth, report endpoints).

Result

Migrated 20 professionals, 224 services and 500 clients to the new schema; cut response latency and cost.

Public chat that qualifies and drops the lead into the panel

Problem

Every clinic needed its own inbound channel, handled with the care health data demands.

Solution

A CRM module inside the platform itself: public chat with an AI agent, lead capture, CFM and LGPD guardrails and consent logging. Per-conversation cost was measured against the API before shipping, to fit the agreed ceiling.

Virtual assistant that answers, qualifies and books

Problem

A service business losing leads to slow first replies, with nobody to book appointments after hours.

Solution

An n8n agent that handles text, audio (Whisper) and images, qualifies, books and retains — presented on a page with a live chat demo wired to the real flow, so the client can try it before buying.

A month of tickets becomes a customer-success report

Problem

Helpdesk ticket volume with no business reading: nobody knew why clients opened tickets or where churn risk was.

Solution

An n8n agent that pulls the month's tickets through the helpdesk API and generates a per-client report — reasons for contact, open/closed/resolved counts, recurring subjects, risks and opportunities. Customer success, not support metrics.

Full personal-finance web app in production

Problem

Scattered personal finance management with no API for automation.

Solution

Full app: recurring expenses, installments, goals, BCB rate sync (CDI/Selic/IPCA), investment simulator, and a versioned REST API with API keys consumed by external n8n automation.

AI SaaS for the veterinary sector

Problem

Veterinary service without a unified intelligent platform.

Solution

Web platform (SaaS) with dedicated AI, ChatGPT/Claude-style interface, Supabase auth/DB/realtime and edge functions (user creation, audio transcription). WhatsApp channel addendum proposed.

Management for 3D printing pros

Problem

3D printing pros without a tool for store, cost and order management.

Solution

Management PWA: store, catalog, orders, inventory and cost calculation, with Stripe payments and NextAuth v5 auth.

The client approves by link and the app publishes on its own

Problem

Content approval over WhatsApp and email: no history, no deadline, and manual publishing on every network.

Solution

Multi-tenant SaaS where each client reviews posts through a public link and, on approval, the app schedules and publishes via the Meta Graph API. Separate agency and client roles, per-project RLS and an auditable approval trail.

Real-time funnel of AI-handled conversations

Problem

Managers had zero visibility into what the AI agents were doing in the funnel — conversations, qualifications and follow-ups were invisible.

Solution

Dashboard fed by n8n workflow events through an edge function authenticated with a per-client token (lead_created, conversation_started, lead_qualified, followup_sent, deal_won…). KPIs, conversion funnel, per-lead timeline, sales leaderboard and goals with celebrations.

Result

Generic event contract: each new agent plugs into the dashboard with 3 emitter workflows pointing at the same endpoint.

End-to-end GTM automation on Pipedrive/Kommo

Problem

Sales team lost pipeline due to missing cadence, follow-up and post-call context.

Solution

Stage-based activity cadence with rollback, AI follow-up, Whisper call summaries logged to deals, candidate screening into ClickUp, and Meta Lead Ads ingestion with segmented WhatsApp dispatch.

Semantic search with AI visual verification

Problem

3D model archives scattered across ~60 Telegram groups, impossible to search.

Solution

Search engine fusing fuzzy + vector search over 1,266 indexed libraries, with GPT-4o Vision visual verification before confirming a match and on-demand download.

One selfie returns every photo you appear in

Problem

A wedding archive of nearly 4,000 photos on Dropbox: nobody finds their own without scrolling the whole album.

Solution

Archive indexed with face detection and embeddings (ONNX Runtime on GPU), similarity search in pgvector with an HNSW index, and a mobile PWA where the person takes a selfie and downloads a ZIP. The selfie is never stored and consent is asked before the scan.

Result

3,995 photos and 15,681 faces indexed in ~20 minutes, zero errors, sub-second search. Live in production.

Capture pages that score and route leads

Problem

Cold leads entering the pipeline without prior qualification.

Solution

Capture pages with conditional routing, Meta Pixel and webhook-driven asset delivery — qualifies by revenue and fit before sales (qualified → booking, unqualified → nurture).