Hey,
Looking for freelance n8n projects. Based in Europe, fully remote.
I’ve been building n8n workflows for a while and recently put together a portfolio - four production systems, each solving a different problem with a different architecture. Not templates or tutorials, these run on real data.
Lead Intelligence System - 16 nodes, 3 AI calls
Webhook catches an incoming lead, JavaScript validates it against 20 rules, figures out if the email is business or personal, scores it 0-100 across 12 factors. Then three sequential calls to Groq API - first one analyzes what the person actually wants, second researches their company from the domain, third writes a personalized outreach message. Switch node routes everything by priority - P0 through P3, each with its own SLA and team assignment. Full audit trail, JSON response back to the caller. Eight seconds start to finish.
Built with: Webhook, JavaScript Code nodes, REST API (HTTP Request POST), Groq LLM, Prompt Engineering, Switch routing, Telegram Bot API, Respond to Webhook, Error Handling with retry
Market Monitor - 16 nodes, fully autonomous
Schedule Trigger fires every 12 hours. Pulls articles from three RSS feeds - HackerNews, TechCrunch, The Verge. Each article goes through an NLP engine I wrote in JavaScript - five keyword groups with different weights, title matches score 3x higher than body. Fuzzy deduplication catches the same story across sources. Everything gets packaged into a prompt and sent to Groq for trend analysis. I get a briefing in Telegram, and if Zapier or Make get mentioned — a separate competitor alert.
Built with: Schedule Trigger, RSS Feed Read, JavaScript NLP engine, IF branching, REST API, Groq LLM, Telegram Bot API, Data Aggregation
Support Bot - 15 nodes, real-time Telegram bot
Telegram Bot Trigger picks up messages, JavaScript parser extracts metadata - language, whether there are URLs or phone numbers in the text. Switch node splits traffic three ways - support requests go to AI, commands like /help get instant responses, feedback gets logged separately. Two AI calls for support - first classifies the ticket across seven categories with four urgency levels and sentiment detection, second generates an empathetic response. SLA matrix I built covers 28 combinations - billing plus critical equals 30 minutes, feature request plus low equals 24 hours. Bot replies to the customer and simultaneously drops a full ticket to the internal team channel.
Built with: Telegram Bot Trigger, JavaScript, Switch routing, REST API x2, Groq LLM, Prompt Engineering, Telegram Bot API x3, SLA computation
Company Research Pipeline - 15 nodes, multi-API enrichment
Webhook receives a domain. Three sequential calls to Google DNS API - MX records reveal the email provider, TXT records expose which SaaS services they’ve verified (found Stripe, Atlassian, Salesforce and 11 others on HubSpot), A records identify hosting from IP ranges. HTTP GET fetches the homepage, JavaScript parses HTML for 22 technology signatures - from React to Google Analytics. Everything merges together, tech scoring algorithm rates the company 0-100, Groq builds a profile with a sales recommendation. Result goes to Telegram and back as JSON through Respond to Webhook.
Built with: Webhook, HTTP Request GET x4, Google DNS API, JavaScript HTML parsing, REST API POST, Groq LLM, Telegram Bot API, Respond to Webhook
All code, documentation, and screenshots are on GitHub: GitHub - penkayone/n8n-automation-portfolio: Production-grade n8n workflow automations — AI lead intelligence, competitive market monitoring, customer support bot, multi-API company research. 4 enterprise workflows, 62 nodes, 9 AI calls, 40+ tech signatures. · GitHub
If you have a project that needs n8n work - I’m up for it. Webhooks, API integrations, AI pipelines, Telegram bots, scheduled automations, data enrichment - that’s what I do every day.
Telegram @antongoloskokov or just reply here.
Cheers