All cases.Every system, taken apart.

Where it started, what changed, the results, and where the same approach fits next. Pick a category or browse them all. Click a card to open the case.

Head of sales dashboard: the week's numbers with trends, and a “Now” panel showing what needs attention right nowClient projectVideo Sales · CRMLeads, calls and payments in one CRM2 weeksfrom project kick-off to launch on the client's server “Team pulse”: the whole team on one screen, with quality, training, CRM mistakes and who needs attentionClient projectVideo Sales · ManagementThe whole sales team on one screen6 → 1six spreadsheets, one dashboard, live in 5 days The website's first screen: what the specialist helps with and how to book a free assessmentClient project Website · AI assistantA speech therapist's website: AI answers parents and books assessments4 daysfrom the first call to the website going live on her own domain How it works: chat and calls → lead scoring → CRM → conversation review → management reportIn-house buildVideo Sales · Voice AIAI in the sales team: leads, calls and reports with no manual work27–33 hfreed up per month at 200 leads (calculated) Cover of an EB-2 NIW strategy: ready for the lawyer's review and for the clientClient project Documents · AI analystA client strategy document in 5 minutes, not a week3–5.5 minper strategy document instead of a week of work How it works: email → AI triage → Slack notification and request log → deadline controlIn-house buildVideo Support · EmailClient emails sort themselves, and response times stay under control~13 hfreed up per month at 200 emails (calculated) The chat on this page: an AI consultant connected to live n8n workflowsIn-house buildVideo This website · Live systemA website that talks to visitors and collects leads5–8 sfor the AI consultant to reply to a visitor A request's path: website → n8n → Telegram and a spreadsheet, in 3 secondsClient project Website · LeadsA law firm's website: every request in Telegram in 3 seconds3 sfrom a website request to the owner's Telegram One client, three channels: wrote on Facebook, continued on Instagram, came back to the website, and the bot remembers what was saidIn-house buildVideo Leads · Omnichannel AIOne client. Three channels. One memory.~30 sfrom a client's message to a ready request in Telegram A client writes in Direct: the bot recognises them, and the owner gets a ready-to-act lead in TelegramIn-house buildVideo Instagram · DirectAn AI assistant for Instagram: DMs answered day and night24/7replies in Direct while the team sleeps or is busy A draft in Telegram, approval buttons and the published postIn-house buildVideo Content · InstagramInstagram on autopilot: via API or a ChatGPT subscription1 tapfrom approval to a live Instagram post Night Temple EN: videos from the first month of automation in one recognisable styleIn-house buildVideo Content · YouTubeA YouTube channel that runs itself5–10 minper video instead of 40–60+ minutes of manual routine (from experience) The manager's live dashboard: who's ready, who the system is guiding, who needs a human and whyIn-house buildVideo Onboarding · Telegram bot and AIFrom payment to the first lesson, with no manual reminders2–3 minfrom the first click in the bot to ready for the lesson, measured on the prototype The building chat: the bot posts when the power goes out (with a forecast) and when it's back; on the right, a /status card and a 40% warningIn-house buildVideo Monitoring · Telegram botPower, battery and lift alerts in the building chat5 minthe bot checks the inverter around the clock, 288 times a day The draft on the left, the same document after the system on the right: the journal's margins, font and style, numbered sections; a citation with no source in yellowClient projectVideo Documents · AI + codeJournal-ready formatting: 3–5 hours → 2–5 minutes2–5 mininstead of 3–5 hours of manual formatting Edits right in Word: a new column the reviewer asked for and renumbered citations, in yellowClient project Documents · AI agentReviewer comments worked into the article in one run1 runinstead of hours of an editor's manual work (per the client) Diagram: text → AI agent → Gemini TTS → FFmpeg → Telegram and Google Drive → YouTubeIn-house buildVideo Content · AI integrationsText → voice → YouTube: the AI does the narration itself~35 sof generation for a 15-minute chapter Example: my own server, with sites and dashboards behind Caddy, n8n workflows, databases and services in Docker, and external APIsIn-house build Engineering · InfrastructureTurnkey infrastructure on your own server39workflows running 24/7 on my own server, a live example

Don't see your task here?

These cases are examples, not a catalogue. Describe your process and I'll propose a system built around it. The review is free.

Get a free audit

I name clients only with their consent. Screenshots show sample or redacted data.

Show me the processyou want off your plate.

Describe the task in your own words. I'll reply personally and tell you what can be automated right away.

  1. You describe the task
  2. I reply personally
  3. You get a written plan
INTAKE WORKFLOW: ACTIVE
Email, Telegram or phone. Any channel that works for you.

I reply personally, usually the same day. This form runs on the same n8n stack I build for clients.