Skip to content
View roman-rr's full-sized avatar
🦜
LangChain / LangGraph / DeepAgents
🦜
LangChain / LangGraph / DeepAgents
77 followers · 6 following

Achievements

Achievement: Galaxy BrainAchievement: Pair ExtraordinaireAchievement: YOLOAchievement: Public SponsorAchievement: QuickdrawAchievement: Starstruckx3Achievement: Pull Sharkx2Achievement: Arctic Code Vault Contributor

Achievements

Achievement: Galaxy BrainAchievement: Pair ExtraordinaireAchievement: YOLOAchievement: Public SponsorAchievement: QuickdrawAchievement: Starstruckx3Achievement: Pull Sharkx2Achievement: Arctic Code Vault Contributor

Highlights

Block or report roman-rr

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
roman-rr/README.md

Hey 👋 I'm Roman — AI Solutions Architect & Full-Stack Engineer

  • 🤖 Building multi-agent AI systems with LangGraph, LangChain, and custom orchestration frameworks
  • 📡 Creator of Trading Skills — AI-agentic crypto trading signals for Claude Code, Codex, Cursor & 30+ agents. Free during beta.
  • 🏗️ Architecting production AI pipelines: speech-to-text, real-time email processing, RAG, vector search
  • 🎓 Microsoft Certified AI Engineer · Multi-cloud: AWS · Azure · GCP
  • 🔭 Creator of Cupertino Pane — UI library with 1K+ GitHub stars
  • ⚡ 20+ years full-stack · Healthcare · Real Estate · Software · Web3
  • 👯 Looking for high-complexity challenges — the kind that need an architect, not just a coder

Pinned Loading

  1. trading-skills trading-skills Public

    17 triggers × 44 algorithms × 3 AI experts — live crypto trading signals for Claude Code, Codex, Cursor & 30+ agents. Free during beta.

    Makefile 86 11

  2. tech-systems/panes tech-systems/panes Public

    🎉📱 Create dynamic modals, cards, panes for your applications in few steps. Any framework and free.

    TypeScript 728 42

  3. focused focused Public

    A tool that helps you stay focused longer and boost your work productivity.

    C++

  4. Stop Using Vision LLMs to Read Tradi... Stop Using Vision LLMs to Read Trading Charts — 4 Models, 2 Vendors, 40 Real Signals: 51% Direction, 1/215 Patterns, 100% Long-Bias on Gemini
    1 
    # Stop Using Vision LLMs to Read Trading Charts — 4 Models, 2 Vendors, 40 Real Signals: 51% Direction, 1/215 Patterns, 100% Long-Bias on Gemini
    2 
    
              
    
              
    
                3
                
    > A deep empirical audit of using frontier vision LLMs as chart-pattern experts in a production crypto signal pipeline. Tested against QuantAgent (arXiv:2509.09995), the paper repeatedly cited as proof that LLM agents can trade from candlesticks.
    
              
    
              
    
                4
                
    >
    
              
    
              
    
                5
                
    > Conducted April 2026. Every model called, every call logged, every fixture tied to a verified trade outcome. Four frontier models, two vendors (Anthropic, Google), 215 LLM calls, $1.16 total spent.
    
              
    
          
    
    
    
  
    

    5. 

      

  5. 
  
    
    
    
      
    
        
        Ruflo / Claude-Flow Audit: 300+ MCP ...  Ruflo / Claude-Flow Audit: 300+ MCP Tools — 99% Theater, 1% Real, 5x Token Waste      
    

          
    
              
    
                1
                
    # Ruflo / Claude-Flow: 300+ MCP Tools Exposed — 99% Theater, 1% Real, 5x Token Waste
    
              
    
              
    
                2
                
    
              
    
              
    
                3
                
    > **A deep technical audit of the ruflo (@claude-flow/cli) ecosystem — what it claims vs what it actually does.**
    
              
    
              
    
                4
                
    >
    
              
    
              
    
                5
                
    > Investigated 2026-04-04 by spawning 8 research agents across two analysis phases, reading source code from the [ruflo GitHub repo](https://github.com/ruvnet/ruflo), tracing local process execution, and testing every major tool category hands-on.
    
              
    
          
    
    
    
  
    

    6. 

      

  6. 
  
    
    
    
      
    
        
        MemPalace Exposed: 42,000 Purchased ...  MemPalace Exposed: 42,000 Purchased Stars, Zero Innovation — It's ChromaDB With a Celebrity Name      
    

          
    
              
    
                1
                
    # MemPalace Exposed: 42,000 Purchased Stars, Zero Innovation — It's ChromaDB With a Celebrity Name
    
              
    
              
    
                2
                
    
              
    
              
    
                3
                
    > **A deep technical audit of MemPalace (github.com/MemPalace/mempalace) — what it claims vs what it actually is.**
    
              
    
              
    
                4
                
    >
    
              
    
              
    
                5
                
    > Every source file read. Every benchmark traced. Every MCP tool verified. Compared against Mem0, Zep/Graphiti, and Letta/MemGPT. GitHub stargazer timestamps analyzed for bot patterns.
    
              
    
          
    
    
    
  
    

    7.