Yevgen`s official homepage
Yevgen Somochkin
"ΞΣΣΘ"
CTO ESSO.DEV
Building the Future of AI, Automation & Web3 | Full-stack & DevOps Architect | Scalable, Data-driven Web & Mobile Apps
Hamburg, Germany
Yevgen`s official homepage
Yevgen Somochkin
"ΞΣΣΘ"
CTO ESSO.DEV
Building the Future of AI, Automation & Web3 | Full-stack & DevOps Architect | Scalable, Data-driven Web & Mobile Apps
Hamburg, Germany
Products I'm building — from idea to production
Multi-tenant platform for deploying AI agents on any website. LangGraph StateGraph engine with intent-aware routing, pgvector knowledge base, built-in CRM, appointment booking, and a 3-layer skill system. Widget embeds in one <script> tag. Powers chatbots for restaurants, clinics, law firms and more across DE/EU.
The AI interface layer for blockchain. Connect any wallet to AI agents without exposing private keys. Swap, bridge, and manage portfolios through Claude, ChatGPT, or any MCP-compatible client.
Monitor how ChatGPT, Claude, Perplexity and other AI models recommend your brand vs competitors. Track AI-driven search visibility across 10+ European languages.
MIT-licensed npm package enabling AI agents to interact with DeFi. 6 read-only tools (get_balance, get_portfolio, get_price, get_quote, get_gas, get_history) and 2 write tools (swap, send). Supports Ethereum, BNB Chain, Arbitrum, Base, and Polygon. Includes a Claude Skill for guided DeFi workflows with smart slippage, gas optimization, and safety checks.
Aircraft spotting game with a 20-rank progression system, rarity classifications, leaderboards, and a database of 300+ aircraft models. Spot planes in the sky, scan them, and build your collection.
Map discovery, community chats, business listings with integrated omnichannel customer management. Built-in CRM with Salesforce & HubSpot sync, real-time messaging, and multi-tenant architecture.
Comprehensive technical SEO analysis — performance scoring, Core Web Vitals, structured data validation, and actionable recommendations powered by AI insights.
2378 contributions in the last year
| Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | Jan | Feb | Mar | Apr | |||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mon | |||||||||||||||||||||||||||||||||||||||||||||||||||||
| Wed | |||||||||||||||||||||||||||||||||||||||||||||||||||||
| Fri | |||||||||||||||||||||||||||||||||||||||||||||||||||||
Recent Blog Posts
Apr 5, 2026
Popular Questions
Use AI-powered news aggregators that filter thousands of daily crypto articles down to the most important ones, set up topic-specific alerts for tokens you hold, and dedicate fixed time slots for news review rather than constant monitoring.
AI filters crypto news by analyzing article content with NLP — scoring relevance based on topic keywords, named entities (tokens, protocols, people), sentiment, source credibility, and freshness. It then deduplicates similar stories so only the most informative version surfaces.
An AI-powered news aggregator uses machine learning and natural language processing to automatically collect, filter, and rank news articles from thousands of sources — delivering only the most relevant content instead of overwhelming users with raw feeds.
Main risks: private key exposure, AI hallucination executing wrong transactions, slippage/MEV exploitation, supply chain attacks from malicious servers, and prompt injection triggering unauthorized transactions. Mitigate with non-custodial architecture, value limits, and verified open-source servers.
It depends on architecture. Custodial mode (server holds private key) is risky — if compromised, funds are lost. Non-custodial (recommended): the server prepares unsigned transactions, you sign via MetaMask or Ledger. Never store private keys in MCP config files; use environment variables and dedicated wallets.
Details
