Vitalii Karasov
Rating
Language proficiency level
CV
I design and build AI-powered automation systems that eliminate repetitive manual work across entire business operations — lead generation and qualification, CRM management, content production at scale, competitive intelligence, sales outreach sequences, customer onboarding, data extraction and enrichment, internal reporting, email marketing, document processing, and cross-platform integrations.
My technical stack: n8n (self-hosted and cloud), Claude Code, Claude API, ChatGPT API, Python, Supabase (pgvector for RAG), Docker, Make.com, webhooks, REST APIs, Google Sheets automation, Airtable, Slack integrations, SMTP/email automation, Telegram bots, database-driven workflows, scheduled triggers, and multi-step error handling architectures.
Before going full-time into AI automation, I spent 4 years managing complex cross-functional projects for clients and teams across the US, EU, and Asia — coordinating stakeholders, building systems from scratch, handling budgets, deadlines, and delivery pipelines in high-pressure environments. Master's degree in Strategic Project Management from EU. PRINCE2 certified. I bring project management discipline into every automation I build — clear structure, documentation, scalability, maintainability.
Production results across dozens of projects: multi-agent research systems that compress weeks of analyst work into minutes. Lead qualification pipelines processing hundreds of incoming leads per day with AI scoring, routing, and CRM logging — zero manual touch. Content generation engines producing tens of posts daily at fractions of a cent per API call, with brand voice control, multilingual output, and automated scheduling. Full CRM automation — from first touch to closed deal — eliminating entire manual workflows that used to require dedicated staff. Sales outreach systems running completely hands-off: prospecting, enrichment, personalized messaging, follow-ups, and response tracking. Data parsing and enrichment from Google Maps, LinkedIn, open databases, company directories — structured into clean pipelines feeding dashboards, CRMs, or direct outreach. Automated reporting systems pulling from multiple sources, generating AI commentary, and pushing formatted reports to Looker Studio, Google Sheets, or Slack channels on schedule.
Every system built production-grade: comprehensive error handling, automatic retries, fallback logic, structured logging, real-time monitoring, alerting on failures, and full documentation so the client can maintain and extend without me. I build systems that run without supervision — not prototypes, not demos, not proof-of-concepts.
Skills and abilities
Services
Outsourcing & consulting
Portfolio
-
Multi-Agent AI Court for bodycam (3rd place, Milan AI Week 2026)
AI & Machine LearningBuilt in 7 days solo for the AI Agent Olympics Hackathon — the official hackathon of AI Week 2026 in Milan (the largest AI event in Europe). 3rd place out of 731 teams and 2382 participants.
Problem: Every year, tens of millions of police interactions are recorded on bodycams in the US and EU. Less than 1% ever undergo an audit. When a complaint comes in months later, the officer's written report de facto becomes the sole source of truth. SENTINEL is an audit layer that has never existed: an adversarial multi-agent "court" that processes each recording, cites specific statements and specific frames, and issues a court-admissible verdict in minutes, not months.
… Architecture:
Three layers with different speeds and depths of analysis:
Layer 1 — Rapid (
-
Market Intelligence System — Full GTM in
AI & Machine LearningAutomated competitive research and go-to-market planning for any product or niche. Form submission triggers 5 sub-workflows running in parallel: input validation, competitive intelligence via Gemini Deep Research (autonomous agent that plans its own searches and returns cited reports), market strategy synthesis, GTM planning, and quality evaluation. Quality scoring includes grounding ratio — measures what percentage of claims in the output are backed by real web sources. Score on last run: 1.0 (100%). Results delivered automatically to Airtable and Slack.
https://www.loom.com/share/6a08c5221efd444db95ee0275247f288
… #N8N #gemini #airtable #aiagent #Slack
-
AI Content Engine — 30 Posts for $1
AI & Machine LearningFull closed-loop content system that writes in brand voice 6 connected Make.com scenarios: INGESTION: loads brand voice, audience pain points, vectors into Pinecone GENERATION: queries RAG vectors, structured prompt to Claude, 6 draft variants INTEGRATION WEBHOOKS: unified pipeline triggerable from Slack forms or other scenarios PUBLISH: pushes approved variant to LinkedIn and X via Buffer FEEDBACK LOOP: collects engagement metrics, auto-adds top posts back to Pinecone as style references REDDIT CHECK: monitors subreddits for fresh audience pain points, updates audience language automatically
https://www.loom.com/share/3e08552f874f407bad4b558bd0bdf9d8
… #make.com #pinecone #claude #promptengineering #Buffer
-
Freelancer Toolkit for Claude Code (Custom Skills & Sub-Agents)
AI & Machine LearningCollection of 5 production-ready Claude Code slash commands: /project-init (scaffolds entire project in 10 sec), /code-review (automated bug & security analysis), /deploy-checklist (pre-deployment verification), /client-report (git log to client-friendly summary), /estimate (task estimation with hour range and pricing). Open-source on GitHub.
https://www.loom.com/share/28366f5e5ffd4ad3a3e7f922e9afcf10
… #claude-code #VisualStudio #authomation #python #GitHub
-
AI Photo Culling & Tagging CLI Tool (Python + OpenAI Vision)
AI & Machine LearningCLI tool for photographers that processes RAW photos, extracts EXIF metadata, generates 512px previews, sends to GPT-5.4 Vision for scoring 1-1000 with reasoning. Processed 121 Panasonic RW2 files in 15 minutes for $0.35. Recommends top photos for Instagram Stories. Open-source on GitHub.
https://www.loom.com/share/adb04c24f46540d8b544de8e488eb88c
… #python #visionapi #openai-api #pillow #claude-code
-
AI Lead Qualification Bot (n8n + Claude API)
AI & Machine LearningAutomated lead classification system. Receives lead data via webhook, Claude API classifies as hot/warm/cold, generates personalized response, saves to Google Sheets. Processes each lead in under 5 seconds, saving ~5 hours/week for 100 leads.
https://www.loom.com/share/9c2ef40dee6543f897060ef4b8596d74
… #N8N #claude #openai #docker #authomation
-
Multi-Agent AI Research System (n8n + Claude + RAG)
AI & Machine LearningBuilt a system with 4 specialized AI agents that replaces 3 hours of manual research with a 2-minute automated process. Supervisor agent coordinates Research, Historian, Strategist, and Narrator sub-agents. RAG pipeline with 160+ document chunks stored in Supabase pgvector.
https://www.loom.com/share/d99029c2ac5b4969823c818cb81ef0d5
… #N8N #claude #openai-api #Supabase #docker
Reviews and compliments on completed projects 1
Vitaliy professionally set up AI automation for processing leads from Calendly. He implemented integration with ClickUp, Gmail, and AI analysis of applications, configured the logic for follow-up emails and lead statuses. He made all necessary adjustments, tested the scenarios, and was always in touch. I am very satisfied with the result and the collaboration. I recommend!