ebartan specializes in developing AI-powered automation solutions and intelligent agent systems using cutting-edge technologies like Hermes, OpenClaw, Paperclip, and modern LLM frameworks. Embracing the Vibe Coding philosophy — a windsurf-inspired, flow-focused approach to effortless, intuitive development with AI.
Experience
Building AI systems and automation solutions for global clients
Upwork · 2023 - Present
Delivering AI integration, workflow automation, and full-stack development solutions for global clients. 30+ successfully completed jobs with 233+ hours billed, $45/hr and 100% Job Success.
Hermes · OpenClaw · Paperclip · 2024 - Present
Building multi-agent systems with persistent memory using Hermes, OpenClaw, and Paperclip frameworks. Cross-session context retention and intelligent task automation. Vibe Coding philosophy — like a windsurfer finding flow with the wind, intuitive and flow-focused development.
Bubble.io · Softr · Make · 2022 - Present
Building responsive web applications and automation pipelines with low-code platforms. Designing end-to-end workflow automations that eliminate repetitive tasks and connect disparate systems.
Solana · Neon · Next.js · 2021 - 2022
Built Web3 applications on Solana and Neon blockchain. NFT catalog platforms with Next.js, wallet connection and authentication systems, decentralized application (dApp) interfaces. Manual full-stack Web3 solutions in an era before AI.
GitHub · 2020 - Present
450+ repositories on GitHub. Contributing to LangChain-Qdrant integrations, AI content pipelines, and workflow automation tools. Building community projects for the AI ecosystem.
Features
Hermes, OpenClaw and Paperclip AI agent orchestration, Neo4j persistent memory, Inngest workflow management, and GraphRAG architecture — solutions to take your business into an autonomous future.
Paperclip, Hermes and OpenClaw orchestration for 24/7 autonomous operations targeting "zero human" intervention business models. Org charts, budgets, and goal alignment with full control.
Inngest "workflow brain" with auto-healing errors, rate limit management, and systems that never stop. Automatic retries, idempotency, and concurrency control.
Neo4j graph database-based GraphRAG architecture delivering more explainable, traceable, and accurate AI responses than traditional RAG. Instant vector search and semantic querying.
Neo4j + Voyage AI-powered agents that remember past conversations and evolve with you. Cross-session context retention, semantic memory, and vector search.
Fivetran ELT/CDC for real-time data synchronization, SingleStore HTAP for millisecond-speed analytical insights, Neon serverless PostgreSQL for auto-scaling storage.
n8n and Flowise automations delivering 30%+ productivity increase on repetitive tasks. Composio with 3000+ tool integrations. Vibe Coding philosophy for fluid and intuitive solutions.
AI Agent Ecosystem
Advanced AI agent that learns skills from experience, working via CLI and messaging platforms. Tool calling, persistent memory, and autonomous execution.
All memory and semantic search managed with Neo4j graph database + Voyage AI embeddings (1024d). Faster, cheaper, full control.
Application reliability layer managing durable, event-driven workflows between systems. Automatic retries, idempotency, and concurrency control for uninterrupted operations.
If OpenClaw is an employee, Paperclip is the company itself. Budget control, goal alignment, and management of different agents (Claude Code, Codex, Cursor) from a single dashboard.
Connect tools, automate repetitive tasks, and build end-to-end pipelines with no-code visual workflows.
Replace manual busywork with automated triggers and actions — your team focuses on what matters.
From simple webhooks to complex multi-step integrations, grow without adding overhead.
Step 1: Define & Design
Choose between Hermes (self-improving agent), OpenClaw (agent gateway), or Paperclip (autonomous company) frameworks based on your needs. Design a persistent memory strategy with the Letta Framework, define the agent's persona and strategic knowledge through SOUL.md.
Step 2: Build & Connect
Connect to 3000+ external tools via Composio, build high-accuracy semantic access with Neo4j GraphRAG. Set up auto-retry, fault tolerance, and event-driven flows across all systems with the Inngest "workflow brain".
Step 3: Deploy & Scale
Launch on serverless architecture with Vercel or Railway. Autonomous orchestration through Paperclip with budget control, goal alignment, and heartbeat mechanisms. All decisions and costs tracked in real-time from the dashboard.
Ready!
<graph>
Graph database relationships used in projects — real Neo4j integrations in production
Ecosystem
Tools and platforms I use daily to deliver AI-powered solutions.
Tools
Modern AI agent frameworks and cloud services for intelligent automation
Hermes Agent — evolving AI that learns skills from experience. OpenClaw — open-source gateway for agent lifecycle and model routing. Paperclip — orchestration, org charts and budget management for autonomous AI companies.
Inngest — event-driven "application reliability layer" with auto-retry and concurrency control. n8n — open-source workflow automation. Composio — 3000+ tool connections (GitHub, Slack, Reddit). Flowise — drag-and-drop LLM flow design.
Fivetran — ELT data integration with 500+ pre-built connectors and CDC real-time sync. SingleStore — hybrid transactional/analytical (HTAP) distributed SQL database. Neon — serverless PostgreSQL, auto-scaling with instant branch support.
Neo4j — graph database for GraphRAG and recommendation systems. Zilliz Cloud — high-scale vector similarity search. Pinecone, Qdrant, Milvus — vector stores for RAG and semantic search.
OpenAI GPT, Anthropic Claude, Groq LPU ultra-fast inference, Hugging Face open-source models. LangChain — LLM application framework. LangGraph — multi-step AI workflows. Voyage AI — vector embeddings (voyage-3-large, 1024d).
xurl CLI — X/Twitter API from terminal (tweet, DM, search, media). Kapso — official WhatsApp Business API. Hermes TUI — CLI-based AI agent interface. Pydantic-AI — data validation and type safety.
Docker — container-based deployment. Vercel — Next.js/Astro frontend hosting. Railway — rapid backend deploy. Cloudflare — DNS, tunnel, CDN. InsForge — Edge Functions + PostgreSQL backend platform.
Articles on AI agents, automation, full-stack development, and lessons learned since 2020.
How I evolved from static Astro sites to AI-powered full-stack applications on Vercel over 1,400 days. Serverless functions, Inngest background jobs, multi-platform strategy with Railway and Cloudflare.
How to make your website discoverable by AI models like ChatGPT, Perplexity, and Gemini. A step-by-step guide to two critical tools of AI discoverability: llms.txt and JSON-LD structured data.
CodeGraph is a pre-indexed code knowledge graph for AI coding agents. AST-level parsing with SQLite-based knowledge graph — works with Claude Code, Cursor, Codex, and OpenCode.
DeepSeek V4 Flash and Pro model features, pricing, 384K context window, and OpenAI/Anthropic compatible API integration. Agent tools and practical usage examples.
FAQs
Quick answers to common questions about my work and process.
Higher accuracy and more complete answers (runtime / production benefit). Once you've created your knowledge graph, it's easier to build and maintain your RAG application. Better explainability, traceability, and access controls (governance benefit).
Railway is a deployment platform that lets you provision infrastructure, develop locally with that infrastructure, and deploy to the cloud or simply run ready-made software from our template marketplace.
AI otomasyonu ile işinizi bir üst seviyeye taşıyın.
