Available for collaborations • Privacy-respecting AI
Marcin Baran
AI/ML Engineer | Full-Stack Developer
Building production-ready AI systems. Career transition from engineering to software development.
Featured Projects
A few production-oriented builds that reflect my current focus.
AstraForge
AI-Powered Monorepo Factory
Revolutionary AI factory that generates production-ready applications across 4 platforms simultaneously using natural language. 5 specialized agents handle architecture, coding, testing, deployment, and monitoring.
Metrics
- • platforms: 4
- • buildTime: 4 minutes
- • agents: 5
- • status: Live demo
Impact
Reduced development time by 80% for full-stack applications, enabling rapid prototyping and production deployment
Key Challenges Solved
- • Multi-agent coordination and state management
- • Cross-platform code generation consistency
CodeSentinel
AI GitHub Repository Analyzer
Professional AI-powered codebase analysis platform for engineering teams. Identifies security vulnerabilities, performance bottlenecks, and technical debt with automated PDF reporting.
Metrics
- • analysis: Real-time
- • security: Enterprise-grade
- • deployment: Production
- • authentication: Clerk + OAuth
Impact
Enabled automated code analysis with AI-powered security scanning, helping teams identify vulnerabilities and technical debt efficiently
Key Challenges Solved
- • Real-time repository analysis and processing
- • Security vulnerability detection accuracy
Claude RAG Chatbot
Production Document Intelligence with Claude
Enterprise-grade RAG system powered by Claude Sonnet 4. Upload documents and get accurate answers with source citations.
Metrics
- • model: Claude Sonnet 4
- • formats: 4 types
- • deployment: Production
- • features: Real-time chat
Impact
Enabled enterprise teams to query 10K+ documents with <2s response time, improving information retrieval efficiency by 90%
Key Challenges Solved
- • Real-time streaming responses
- • Multi-format document parsing accuracy
Enterprise RAG System
Production Document Intelligence
GDPR-compliant RAG platform with local LLM deployment for enterprise document analysis. Designed for privacy-conscious organizations requiring on-premise AI solutions.
Metrics
- • deployment: On-premise
- • compliance: GDPR-ready
- • architecture: Local LLM
- • formats: Multi-format
Impact
Built GDPR-compliant RAG system with local LLM deployment, enabling on-premise document intelligence with zero data leakage and significant cost savings vs cloud solutions
Key Challenges Solved
- • Local LLM deployment and optimization
- • Achieving <2.1% hallucination rate