Eka AI Debugger

🚀 AI-powered debugging platform for developers and teams Analyze code, logs and stack traces using multiple AI providers and turn solutions into a reusable knowledge base

Eka AI Debugger is a production-grade, multi-tenant debugging and analysis platform designed for real-world development teams, SaaS companies and technical support environments.

It allows you to analyze errors using both cloud and local AI models, detect recurring issues and build a persistent internal knowledge base.

---

⚡ Core Features

• Multi AI Provider Support (OpenAI, Anthropic, OpenRouter, NVIDIA, HuggingFace)
• Local Model Integration (LM Studio, Ollama)
• Debug Sessions with code, logs and stack traces
• Similar Error Detection (historical matching)
• Knowledge Base from resolved issues
• Token & Cost Tracking Dashboard
• Multi-tenant workspace system
• Premium SaaS dashboard interface

---

🧠 Why this project?

Most AI debugging tools are either limited, expensive or not self-hostable.

Eka AI Debugger provides:

• Full control over your AI providers
• Local model support (no API cost required)
• Reusable internal knowledge base
• Multi-tenant architecture for real SaaS usage

---

📸 Screenshots

Login	Debug Session

Settings	Knowledge Base

---

🧪 Example Analysis Output

{
  "root_cause": "Null reference in database connection",
  "severity": "high",
  "suggested_fix": "Check database connection before query execution",
  "optimization": "Use connection pooling",
  "security_note": "Avoid exposing database errors in production"
}

---

🏗️ Architecture Overview

Multi-Tenant System

The platform uses a workspace-based architecture where all debug sessions, logs and knowledge entries are isolated per tenant.

AI Provider Layer

The system supports both cloud and local providers:

• OpenAI
• Anthropic (Claude)
• OpenRouter
• NVIDIA AI
• HuggingFace
• LM Studio (local)
• Ollama (local)

Provider-based architecture allows flexible switching and scaling.

Debug Engine

• Accepts code, logs, stack traces and notes
• Performs structured AI analysis
• Stores results for future reuse
• Detects similar historical issues

---

🛠️ Tech Stack

• Backend: Python (FastAPI)
• Frontend: Tailwind CSS (premium SaaS UI)
• Database: MySQL / SQLite
• AI Integration: OpenAI, Anthropic, HTTPX

---

⚙️ Installation

Requirements

• Python 3.10+
• MySQL or SQLite

---

Setup

cd eka-ai-debugger
pip install -r requirements.txt

Create .env file:

DATABASE_URL=mysql+pymysql://root:password@127.0.0.1:3306/eka_ai_debugger
SECRET_KEY=your-secret-key
ACCESS_TOKEN_EXPIRE_MINUTES=1440

Run server:

uvicorn main:app --host 0.0.0.0 --port 8000

---

🔐 Demo Access

Admin info@ekayazilim.com.tr / ekasunucu

---

💼 Use Cases

• Debug production errors quickly
• Analyze logs with AI assistance
• Build internal debugging knowledge base
• Reduce repeated debugging effort
• Improve developer productivity

---

🛣️ Roadmap

• Advanced AI error clustering
• Team collaboration tools
• Export reports (PDF / JSON)
• Webhook & alert system
• CI/CD integration

---

⭐ Support

If you find this project useful, consider giving it a ⭐ on GitHub.

---

💼 Commercial Use

Custom SaaS development, AI integrations and enterprise solutions available.

📧 info@ekayazilim.com.tr

---